Habitat preference and population density of threatened Visayan hornbills Penelopides panini and Rhabdotorrhinus waldeni in the Philippines

: The habitat preference and population density of the Critically Endangered Rufous-headed Hornbill Rhabdotorrhinus waldeni and Endangered Visayan Tarictic Hornbill Penelopides panini were studied in the Central Panay Mountain range in Panay Island, western Visayas. Point counts were established to survey for hornbills and habitat variables were measured in a 30 x 30 m plot at each point. P. panini was recorded in primary and secondary forest, while records of R. waldeni were mostly in primary forest, with only one record in secondary forest. Spearman’s rank order correlation and stepwise multiple regression showed that R. waldeni is positively correlated to density of large trees (>70cm DBH), with the Visayan Tarictic Hornbill showing moderate correlation with density of medium-sized trees (31–70 cm DBH). Central Panay Mountains is the largest remaining forests block in western Visayas. It is the stronghold area of R. waldeni and other western Visayas threatened species, supporting the need to declare the site a protected area.


INTRODUCTION
There are 62 species of hornbills in the world, of which 10 species and six subspecies are endemic to the Philippines (Gonzalez et al. 2013;Poonswad et al. 2013). All Philippine hornbills occur in lowland forests below 1,200 m (Gonzalez et al. 2013), but forest loss in the lowlands may have pushed species to the upper limit of up to 1,500 m elevation (Poonswad et al. 2013). Of the 10 species in the Philippines, two are classified as Critically Endangered, two as Endangered, and six as Vulnerable (IUCN 2020). The Ticao Hornbill Penelopides panini ticaensis, subspecies of the Visayan Tarictic Hornbill was declared Extinct (Poonswad et al. 2013) because of habitat loss. The Rufous-headed Hornbill Rhabdotorrhinus waldeni and P. panini are restrictedrange species found only in the islands of Negros and Panay in western Visayas, where they inhabit lowland forest (Image 1 & 2). With continuing lowland forest loss, populations have been assessed by BirdLife and the International Union for the Conservation of Nature (IUCN) as decreasing. BirdLife International (2020) and the IUCN (2020) estimate the Panay population at 1,800 individuals, with 1,200 mature individuals, and declared the Negros population as possibly extinct.
Currently, only three protected areas (PAs) exist in Panay: Northwest Panay Peninsula Protected Landscape, Sibalom Natural Park, and Bulabog-Putian Natural Park. Despite ongoing conservation efforts, the largest remaining forest and stronghold of hornbill populations is in the Central Panay Mountains, and remains unprotected. This paper presents information on the habitat, abundance and conservation status of P. panini and R. waldeni in CPM. Conservation initiatives primarily led by local government units and Department of Environment and Natural Resources (DENR) are also presented. This research was undertaken as part of the Philippine Hornbills Conservation Programme of Philippines Biodiversity Conservation Foundation Inc.

Study area
The Central Panay Mountain (CPM) refers to the chain of mountains that stretches and divides Panay Island into four provinces: Iloilo, Antique, Aklan, and Capiz (Klop et al. 2000;Mallari et al. 2000

Mt. Camantra (11.544778°N & 122.148408°E)
The Mt. Camantra is located in the municipality of Sebaste (Figure 1) with elevation range of 444-996 m. The primary forest covers steep ridges and gullies. The average height of trees was 25 m and majority have more than 30 cm diameter at breast height. Canopy cover ranges 60-100% forest cover. Thick and tall undergrowth cover the trail and some areas were almost inaccessible with fallen dead trees and decaying logs. Forest trails were hardly visible indicating limited human-related activities. The secondary lowland forest habitat (11.544139° N & 122.147083° E) was located at 419 m. Majority of the trees have heights ranging 10-15 m with average diameter at breast height of 30 cm (n= 41 plots). The trails were visible following rivers, streams, and mountain ridges. The canopy cover ranged from 40-80 %. Abandoned agricultural clearings and presence of pineapple and coconut plantations were observed in the lower sections of the forests closer to the town and lower reaches of rivers and streams.

Mt. Igpako (11.466167°N & 122.1355°E)
The Mt. Igpako site is located in the municipality of Culasi ( Figure 1) with primary lowland forests reaching up to 631 m. The forest understory was thick with Pandanus sp. and few climbing bamboo species. Large uprooted trees from recent typhoons were encountered. Canopy cover ranges 60-90% with patches of openings created from uprooted trees. In elevations above 900 m, moss cover thickly covered the forest floor and barks of trees.

Hornbill distribution and abundance: point-count sampling
Twenty-four-point count stations were established and surveyed, 15 located in the primary lowland forests of Mt. Camantra, Sebaste (5 points) and Mt. Igpako, Culasi (10 points), and nine in the secondary lowland forest in Mt. Camantra, Sebaste. Each sampling point was at least 250 m apart and the observation record bands are 100 m radius. Bird observations were conducted from 0600 h to 1100 h. Two to three observers recorded the distance of the hornbill from the center of the point station as well as the number of individuals seen or heard. Distance of the hornbill was recorded only through estimate. Observation per point was around 10 minutes. 8 x 42 and 10 x 42 roof-type binoculars were used during the survey. We identified forest type following the definition from the Convention on Biological Diversity (CBD): Primary forest is a forest that has never been logged and has developed following natural disturbances and under natural processes. Secondary forest is a forest that has been logged and has recovered naturally or artificially.

Habitat assessment
Twenty-four 30 x 30 m plots were established in each of the 24 sampling points to determine habitat characteristics (Table 1). Canopy cover was measured using an improvised densiometer, and a tape measure was used in acquiring diameter at breast height (DBH). Understory cover percentage was measured through visual estimate. Garmin GPS was used to measure elevation and coordinates.

Data analyses
Hornbill density was acquired by calculating the average number of hornbills per point and dividing it by the total area sampled (3.14 hectares/point * number of point count stations) (Raman & Mudappa 2003). Density and population estimate using DISTANCE program was not used due to insufficient number of encounters for both species of hornbill.
Mann-Whitney U-test was used to compare the habitat variables between primary and secondary forest. Spearman's rank-correlations and multiple linear regression (stepwise) were used to identify relationship between habitat variables and hornbill abundance.

Hornbill distribution and records
We recorded a total of 12 detections (22 individuals) of P. panini and 31 detections (94 individuals) of R. waldeni. P. panini was recorded four times in primary forest and eight times in secondary forest while 99% of the records of R. waldeni were in primary forest. There were 13 R. waldeni positively identified as female and 18 as males and the rest were heard in groups. The highest number of individuals observed in a cluster was 25 (eight females, 13 males and four immature) while the average number of individuals encountered in a flock was 10. In the case of P. panini, three were positively identified as males and four were females.
We observed R. waldeni in secondary forests feeding on fruiting native trees together with P. panini, Pink-bellied Imperial Pigeon Ducula poliocephala, Green Imperial Pigeon Ducula aenea, Philippine Cuckoo dove Macropygia tenuirostris and other fruit-eating species.

Hornbill habitat preference
The primary lowland and secondary forests in CPM varied across sites. There were more large and tall trees in the primary and secondary forests of Mt. Camantra in Sebaste than in Mt. Igpako in Culasi. We tested for significant differences between primary and secondary forests using Mann-Whitney U-tests. Trees with a range of 16-20 m tall and trees with 31-70 and >70cm DBH varied significantly between primary and secondary forests (Table 1; Figure 2). This difference influenced the abundance of the hornbills. The R. waldeni was significantly correlated to the density of trees with 31-70cm DBH (r s = 0.614, P= 0.001) and trees greater than 70cm DBH (r s = 0.618, P= 0.001). This species was recorded almost strictly to primary forest with only one record in secondary forest. There was no significant correlation between habitat variable and P. panini, abundance. P. panini was recorded in both primary and secondary forest with most of the records in the latter.

DISCUSSION
The forests in Sebaste and Culasi still contain significant lowland tropical rainforests. Both hornbills were encountered in primary and secondary lowland forest. Undisturbed forests with tall and large trees were waldeni at 1,000-2,499 mature individuals with total number of individuals at 1,500-4,000. From this study, we acquired a density of 0.2 individuals/hectare (20 individuals/km2) for the P. panini and 0.47 individuals/ hectare (47 individuals/km2) for the R. waldeni. This is a crude estimate from our field observations and basic density calculation from number of recorded hornbills and the sampled area and accuracy can be improved with more data particularly on adjacent forest areas and to run the analysis in the DISTANCE program.
Mynott et al. in 2021 estimated the population density of P. panini in Northwest Panay Peninsula Natural Park (NWPPNP) through DISTANCE analysis to be at 17.8 individuals/km 2 (0.178 individuals/ha) in primary forests and 3.7 individuals/km 2 (0.037 individuals/ha).   (2018)

J TT
Though we were not able the acquire density estimates using DISTANCE analysis and per forest type our crude estimate of 20 individuals/km 2 (0.2 individuals/ha) was close with their primary forest density estimate. In terms of encounters between forest types, we detected the P. panini more in the secondary forest (8 encounters) than the primary forest (4 encounters) while Mynott et al. in 2021 encountered more P. panini in primary forest (31 encounters) than in secondary forest (12 encounters). We suspect that this was the case for the sites we surveyed in CPM because of the presence of R. waldeni, which we observed to be more abundant and dominant particularly in the primary forests. R. waldeni is almost certainly extirpated in NWPPNP (Birdlife International 2022) and its presence needs to be confirmed. With that, there is lesser competition for space and resources for the P. panini in NWPPNP, most significantly in the primary forests. Though the two species of hornbills co-exist and form flocks, there will be competition for nesting trees during breeding season. The P. panini having a broader habitat preference, requiring less pristine conditions than the R. waldeni, were able to occupy the secondary forests if competition for resources ensues. P. panini shows preference to areas with mediumsized trees (30-70 cm DBH) which could be one of their minimum requirement for nesting. This was similar to the results of Klop et al. (2000) in their study of nest site characteristics, where the mean DBH of nesting trees of P. panini was 34 cm. Klop et al. (2000) only measured one nesting tree of R. waldeni that was 90 cm. These results indicate that P. panini has a wider range of habitat preference, and thus it is able to survive in secondary forests and tolerate some level of disturbance. R. waldeni on the other hand has a narrow preferred habitat range, requiring large trees and undisturbed forests. Among the sites in Panay, CPM holds the largest lowland forest cover in Panay followed by Northwest Panay Peninsula Protected Landscape and Sibalom Natural Park.

Hornbill distribution
The past distribution of P. p. panini was on Negros, Panay, Masbate, Sicogon, Pan de Azucar and Guimaras Islands. However, it is now extinct on the latter three and only survives in small forest fragments on Negros and Panay (Poonswad et al. 2013;Paguntalan et al. 2002Paguntalan et al. , 2004

Threats
The primary lowland forests in Sebaste and Culasi are still relatively undisturbed. Part of the reason for this is the topography of the area. The forests are found on steep mountain ridges and gullies that are difficult to access. Trails follow the rivers and streams and bisect steep slopes to reach the mountain ridges. The lower reaches and areas close to the villages showed evidence of disturbances while the forest interior remained relatively pristine. As one moves farther away from the villages and rivers, less disturbance in forests was encountered.
In areas close to the villages and towns, habitat destruction and hunting remain as the main threats to hornbill populations. Large trees were observed being selectively logged for timber needed to make boats and houses or sold as lumber in the nearby town. Clearings were then planted with subsistence crops. Locals were also reportedly using marble guns for hunting birds and other animals for protein and for sport.
During the survey, two of the six known nesting trees of R. waldeni in Mt. Camantra were uprooted by previous typhoons. We also recorded two additional J TT nesting holes suspected to be that of R. waldeni based on the size of the tree, location and size of the nesting hole and the larger seeds found at the base of the tree trunk directly below the nest hole. Locals have also reported this nest as that of R. waldeni.

Conservation efforts
The national threatened status of R. waldeni remained at Critically Endangered status, but P. panini was changed from Endangered to Critically Endangered status in the recent revision of the National List of Threatened Fauna and their categories in the Department Administrative Order 2019 -19 (BMB DENR 2020).
The enactment of the Expanded National Integrated Protected Areas System (ENIPAS) Act of the Philippines effectively legislated into law the declaration of two Protected Areas in Negros (Balinsasayao Twin Lakes Natural Park and Northern Negros Natural Park) and two in Panay (Northwest Panay Peninsula Protected Landscape and Sibalom Natural Park). Mt. Kanlaon Natural Park has its own Republic Act declaring it as a protected area in 2002.
The DENR Region VI initiated the protected area suitability assessment (PASA) last December 2020-January 2021 as the first step in declaring Central Panay Mountain Range as a protected area. This was in response to the resolution passed by the Regional Development Council of Region VI in 2018. There were several attempts in the past to declare the CPMR as a nationally declared protected area, but only local legislations were enacted.

Declaration of Central Panay Mountains as Protected
Area -The Regional Development Council of Region VI and the Department of Environment and Natural Resources (DENR) Region VI had initiated the process of declaring Central Panay Mountains as a protected area last October 2018. The move to declare CPM as PA started in the early 1990s and with millions of funds poured in the last two decades, this has remained a suggestion. Local Government Units (LGU) of all four provinces should seriously investigate securing the very mountains that provided its life support system: freshwater.
Collaboration with LGUs and stakeholders in monitoring protected areas -average annual national budget allocations for monitoring PAs is US $581 while class A local government units have annual funds of at least US $19,380 for environmental programs including monitoring PAs. The Protected Areas Management Board of each PA, DENR and LGU within Key Biodiversity Areas (KBAs) should institutionalize the conduct of synchronized bio-monitoring and hornbill count in each Protected Area and KBA using improved line-point and DISTANCE analysis.
Surveys on threatened hornbills of western Visayas -Surveys are needed in the other identified forests patches, e.g., Northwest Panay Peninsula Protected Landscape, Sibalom Natural Park, Cuernos de Negros Mountain Range, Sta. Catalina forest, Masbate Island, to determine the presence of surviving populations.
Author Details: Andrew Ross T. Reintar -Filipino. An ornithologist and the Field Operations Officer of Philippines Biodiversity Conservation Foundation. Recently conducted a conservation project on the Critically Endangered Cebu flowerpecker and is currently involved in Sulu hornbill and Tawi-tawi biodiversity conservation, working on population and ecological study, education, capacity development and community-based forest protection. Lisa J. Paguntalan -Filipino. An ornithologist and the Executive Director of Philippines Biodiversity Conservation Foundation. She's been doing conservation work for more than two decades focusing on threatened species and habitats in the Philippines. Currently overseeing the Sulu hornbill and Tawi-tawi biodiversity conservation program of PhilBio and other core conservation programs. Philip Godfrey C. Jakosalem -Filipino. Is an ornithologist and the Field Operations Manager of Philippines Bioversity Conservation Foundation (PhilBio). He's been doing conservation work on for two decades, working on bird conservation and ecology. Currently one of the lead of the Sulu hornbill and Tawi-tawi biodiversity conservation program and waterbirds research of PhilBio. Al Christian D. Quidet -Filipino. Currently a medical student of Xavier University in Mindanao. Previously involved in biodiversity research during his undergraduate thesis with particular focus on Visayan hornbills. Dennis A. Warguez -Filipino. An associate professor of Mindanao State University-Iligan Institute of Technology, is currently enrolled as a Ph.D. student at De La Salle University-Manila. He had recently conducted herpetological surveys organized by Herp Watch Pilipinas in Antique, Samal Island, Mt. Apo Protected Area, and Mt. Hamiguitan Range Wildlife Sanctuary. Emelyn Peñaranda -Filipino. Development Management Officer of the Department of Environment and Natural Resources Region 6. She is now actively engaged in the implementation of conservation and development programs and projects concerning protected areas, biodiversity, forest, coastal and marine and urban ecosystems as well as climate change and gender and development.
Author contributions: ARTR-Manuscript writing. Survey and data gathering. Performed the analysis and map making of the study. Conceptualization of the study. Arrangements and preparations of the survey expedition. LJP-Manuscript writing and review. Conceptualization, development and realization of the study and survey expedition. Fund sourcing for the project. PGCJ-Conducted survey and data gathering. Conceptualization and development of the study. ACDQ-Survey and data gathering. Conceptualization of the study. DAW-Survey and data gathering. Conceptualization of the study. Arrangements and preparations of the survey expedition. EP-Arrangements and preparations of the survey expedition. Manuscript review.

www.threatenedtaxa.org
The Journal of Threatened Taxa (JoTT) is dedicated to building evidence for conservation globally by publishing peer-reviewed articles online every month at a reasonably rapid rate at www.threatenedtaxa.org. All articles published in JoTT are registered under Creative Commons Attribution 4.0 International License unless otherwise mentioned. JoTT allows allows unrestricted use, reproduction, and distribution of articles in any medium by providing adequate credit to the author(s) and the source of publication.