An assessment of the diet of Brown Fish-Owl Ketupa zeylonensis (J.F. Gmelin, 1788) (Aves: Strigiformes: Strigidae) from two localities in the foothills of the Western Ghats of Goa, India

: The Brown Fish-Owl Ketupa zeylonensis is a large nocturnal bird of prey that has a vast distribution range. However, there is a significant literature gap on the ecology of this species in the Western Ghats ecoregion, particularly in regard to its food spectrum. In the present study, we assessed the diet composition of this species in the foothills of the Western Ghats of Goa, India. The diet was evaluated by analysing the undigested prey remains in regurgitated pellets obtained from the banks of forest streams and roosting sites. A total of 104 pellets were collected from two localities that exhibited similar landscape characteristics. Our analysis indicated that crabs contributed to a significant proportion of the diet of the species (75.47%), followed by amphibians (frogs, 8.02%), fishes (7.08%), reptiles (snakes, 2.83%), birds (2.36%), scorpions (1.89%), and insects (Odonata, 0.47%). Additionally, 1.89% (n = 4) of the prey items could not be identified due to their disintegrated nature. Furthermore, an assessment of Food Niche Breadth (FNB) indicated that K. zeylonensis exhibited a high degree of specialization in terms of its diet in the study areas.


INTRODUCTION
Birds of prey occupy the apex position in food web assemblages. Therefore, they are considered to be important bioindicators of the environments in which they persist (González-Rubio et al. 2021). The taxonomic order Strigiformes is represented by 250 extant species of owls distributed across the world (Gill et al. 2023). This order is divided into two families: (i) Tytonidae, which includes barn owls, bay owls, and grass owls, and (ii) Strigidae, which includes true (or typical) owls (Sieradzki 2023). India is home to 32 species of owls, 13 of which are found in the state of Goa (Baidya & Bhagat 2018; BirdLife International 2020). The Brown Fish-Owl Ketupa zeylonensis is a nocturnal bird of prey that is distributed across southern and southeastern Asia with isolated populations occurring in Turkey and Iran, and vagrant populations occurring in Seychelles (Birdlife International 2016). It is a large bird (approx. 56 cm) having bright yellow eyes and outward-facing ear tufts. It exhibits rufous-brown upper parts with heavy streaking, and pale underparts with dark streaks (Ali 2002;Kazmierczak & Perlo 2012;Grewal et al. 2016). The species is classified as 'Least Concern' in the IUCN Red List of Threatened Species. Although global populations of this species have not been evaluated, it is suspected to be in decline due to habitat destruction (Birdlife International 2016). In addition, the species is listed under 'Schedule I' of the Indian Wild Life (Protection) Amendment Act, 2022 and under Appendix II of CITES (Ministry of Law and Justice 2022; CITES 2023). In India, this species faces threats from the illegal wildlife trade, persecution by fishermen, and its use in witchcraft (Ahmed 2010).
The Brown Fish-Owl inhabits deciduous, semideciduous and evergreen woodland ecosystems and is found in close proximity to water bodies. Its diet is reported to constitute crabs, fish, frogs, reptiles, birds, mammals, and carrion (Ali 2002;Bindu & Balakrishnan 2015;Grewal et al. 2016).
Owls are highly specialized hunters that regurgitate undigested prey remains such as bones, feathers, hair, scales, and other exoskeletal structures of their prey in the form of compact pellets. The analysis of regurgitated pellets has proven to be a robust technique to assess the food spectrum of owls and understand the diversity and population structure of prey species (Meek et al. 2012;Andrade et al. 2016). In an Indian context, published literature on the diet composition of the Brown Fish-Owl is sparse. Vyas et al. (2013) reported the food spectrum of K. zeylonensis from Jambughoda WS in Gujarat. However, there is a significant literature gap in the diet composition of the species from the Western Ghats ecoregion, particularly in the context to the Indian state of Goa. This study was carried out to understand the diet composition of the species in two sites located in the foothills of the Western Ghats of Goa.

Study Area
Goa is located on the western coast of India (15.492°N, 73.826°E) (Figure 1). The Western Ghats is a 1,600 km long mountain range that runs parallel to the western coast of the Indian peninsula and extends through the Indian states of Gujarat, Maharashtra, Goa, Karnataka, Tamil Nadu, and Kerala. These mountains are recognized as one of the world's eight 'hottest hotspots' for biological diversity and endemism (Molur et al. 2011;UNESCO 2023). In Goa, these mountains pass through the eastern regions of the state where a significant section of the range is protected through four protected areas: Mhadei WS, Bhagwan Mahavir WS & NP, Netravali WS, and Cotigao WS. The vegetation type of the Western Ghats of Goa is varied and includes tropical evergreen, semi-evergreen, and moist mixed deciduous forests (Goa Forest Department 2023). This study was conducted along forest streams that originate from the Western Ghats. The sections of the streams surveyed for this study were located outside the boundaries of protected areas. Study Area 1 was located near Mhadei WS and Study Area 2 was located near Bhagwan Mahavir WS & NP. The streams that were considered for the study were of the perennial and intermittent type. The general vegetation type of the study areas is dominated by tropical evergreen, semi-evergreen, and moist mixed deciduous forests. In addition, both study areas were located in close proximity to plantations and human settlements. The streams considered for this study are part of a larger catchment system that empties into the Mahadayi River of Goa (see Figure 1). The aerial distance between the two study areas was approximately 16.7 km.

Data Collection
This study was conducted from 20 October 2022 to 5 February 2023. Prior to this study, Brown Fish-Owl activity in Study Area 1 was established by conducting field surveys. In addition, the feeding and breeding activity of this species in Study Area 2 was recorded for over two years with the help of camera traps and J TT direct observations respectively. This was part of a larger nocturnal wildlife monitoring effort by Planet Life Foundation, Belloy, Nuvem, Goa and Nature's Nest Nature Resort, Surla, Sancordem, Goa. Brown Fish-Owl pellets were collected from the study areas once a week. The pellets were usually found deposited along stream banks and in close proximity to roosting sites (Image 1). A total of four roosting sites were identified across our study areas based on repetitive pellet deposition observed during our surveys. The entire pellet was collected and temporarily stored in plastic zip-lock bags. Prior to analysis, we manually removed all conspicuous debris from the pellet by hand. Following this, the pellets were soaked in 70% ethyl alcohol for 24 h to kill all microorganisms. The pellets were then air-dried for 24 h to remove moisture. During analysis, the dry weight of each pellet was recorded using a weighing balance with 0.001 g accuracy. The prey items in the pellets were then sorted into eight categories: crabs, insects, scorpions, fishes, amphibians, reptiles, birds, and unidentified prey. These prey categories were established based on literature review and field observations. As we did not have access to reference specimens, the items in the pellets could not be identified at the species level. Identification of the prey items was carried out using reference books and taxonomic keys (Verma 2014;Ganguly et al. 2015;Saxena & Saxena 2019;Mehta et al. 2020;Mishra et al. 2021).
As most of the items in the pellets were conspicuous, identification and sorting were possible by the naked eye. However, we used a compound microscope (ESAW SM-02, ESAW India, Ambala Cantt, Haryana, India) set at 10x magnification to identify the inconspicuous prey remains. Arthropods were identified primarily from structures such as mouthparts, chelipeds, pereiopods, abdomen, and carapace (in the case of crabs); wings (in the case of insects); pedipalps, cephalothorax shield, mesosoma, metasoma, walking legs, and telson (in the case of scorpions). Scorpion identification was also aided by shining an ultraviolet light at 395 nm and observing fluorescence (Gaffin et al. 2012) (Image 1). Chordates were identified from endoskeletal and exoskeletal structures such as bones and scales (in the case of fishes), bones and mouthparts (in the case of amphibians i.e., frogs), vertebrae, ribs, and skin (in the case of reptiles i.e., snakes), and bones and feathers (in the case of birds). The prey items that could not be identified were sorted into the 'unidentified' category. For each pellet, we estimated the number of individuals for each prey category (Table 1). The data for both study sites was pooled and subsequently analyzed.

Data Analysis
We estimated the Relative Frequency of Occurrence (RFO%) for each prey group by dividing the number of occurrences of each prey category by the total number Where P i is the proportion of i th prey category and n is the number of prey categories recorded in the diet of the Brown Fish-Owl. This standardized index computes a value that can range from 0-1. Values closer to 0 indicate a specialist diet whereas values closer to 1 indicate a generalist diet (Mehta et al. 2018).

RESULTS
A total of 104 Brown Fish-Owl pellets were collected during the present study (50 pellets from Study Area 1 and 54 pellets from Study Area 2). The average dry weight of the pellets was estimated to be 4.053 g (SD = ± 2.627; Range = 0.590-12.953). The total number of prey individuals recorded was 212. The average number of prey individuals per pellet was estimated to be 2.029 (SD ± 1.074; Range = 1-5). The diet of the Brown Fish-Owl was dominated by crabs followed by amphibians (frogs), fishes, reptiles (snakes), birds, scorpions, and insects (Odonata). The unidentified prey individuals constituted a minor portion of the diet (n = 4, Table 2). Although we were unable to positively identify the type of prey items in the 'unidentified' category due to their disintegrated nature, we were able to identify the remnants as vertebrates. In such cases, all the unidentified remains having similar characteristics were assumed to originate from a single individual. The number of occurrences of prey categories was largely comparable across the two study areas. However, insects were only present in the pellets collected from Study Area 1 (Odonata, n = 1) and scorpions were only present in pellets collected from Study Area 2 (n = 4) ( Figure 2). Lastly, the Food Niche Breadth (FNB) value was estimated to be 0.1, indicating that the Brown Fish-Owl exhibits a high degree of specialization in terms of its diet in the study areas. The diet composition of the species in the present study has been detailed in Table 2.

DISCUSSION
The Brown Fish-Owl is a nocturnal predator that is known to feed on a wide variety of prey, such as fish, frogs, crabs, small mammals, birds, and reptiles. It is also reported to occasionally feed on carrion (Ali 2002). Published literature on the diet composition of K. zeylonensis in India is sparse. A study conducted by Vyas et al. (2013) on the breeding behaviour of K. zeylonensis in Jambughoda WS and surrounding areas in Gujarat, India reported fishes, crabs, insects, and prawns in the pellets of the species. However, the authors identified several other prey groups such as amphibians, reptiles, and birds from direct feeding observations and analysis of discarded prey items at the nests. This indicates that pellet analysis when supplemented with other observational protocols can significantly aid in the understanding of the food spectrum of the species. The diet composition of K. zeylonensis in Jambughoda WS was very similar to our observations in the Western Ghats of Goa with minor differences (Figure 3). In addition, the study in Jambughoda Wildlife Sanctuary was conducted during the pre-monsoon season (March-April) as compared to the present study that was conducted during the post-monsoon and winter seasons (October-February). Furthermore, fish owls are specialist birds of prey that have preferences for certain prey groups (Sieradzki 2023). Our data analysis supports this fact as the food niche breadth assessment indicated that K. zeylonensis is a specialist predator that feeds mainly on crabs whilst supplementing its diet with other invertebrate and vertebrate prey groups (Figure 2; Table 2).

Crabs
Mouthparts, chelipeds, carapace, abdomen The number of duplicates of exoskeletal structures (either whole parts or fragments) was used to estimate the number of individuals in each pellet.

Fishes
Parts of the axial skeleton (skull, vertebrae, and ribs), and scales.
Microscopic examinations of the morphological patterns on fish scales were conducted based on the principle that the patterns serve as useful taxonomic identifiers of fish species (Bräger & Moritz 2016). This was further supported by observations of the bones from the axial skeleton. As it was difficult to determine the number of individuals of the same species, all endoskeletal remains of similar size were assumed to be derived from a single individual unless morphological examinations of the scales indicated more than one species in the pellet.

Reptiles (Snakes) Vertebrae, ribs, and skin
Identifying the number of individual snakes was straightforward in instances where the vertebral column was found to be relatively intact in the pellets. However, in instances where the vertebral column was found to be in a dismantled state, we used the general shape and size of the vertebrae and ribs to estimate the number of individuals. This was further supplemented by the remnants of snake skin present in the pellets.

Birds
Parts of the endoskeleton and feathers.
The number of duplicate endoskeletal remains was utilized to estimate the number of individuals. In cases where only feathers were present, feathers having similar morphological characteristics were assumed to originate from a single individual.
Pellet analysis is considered to be a robust indicator of the food spectrum of owls. In addition, such analysis can shed light on the richness, evenness, and abundance of prey groups constituting owl diet in the foraging environments (Heisler et al. 2015;Andrade et al. 2016). The present study was conducted due to the gap in knowledge in regards to the diet composition of K. zeylonensis in the Western Ghats ecoregion, particularly in the state of Goa. However, it is imperative to note that pellet collection in the present study was conducted for a relatively short period of time (post-monsoon and winter seasons), and the diet composition of owls is reported to change based on seasonal variations in prey availability (Kafkaletou-Diez et al. 2008;Santhanakrishnan et al. 2010). This may be an important factor to consider in landscapes such as the Western Ghats that undergo changes in hydrology across seasons. Organisms in such aquatic ecosystems may exhibit population changes on a seasonal scale that may influence the diet composition of the Brown Fish-Owl. Therefore, further assessments are required to understand the trends in the diet composition of the species across a seasonal gradient in the Western Ghats landscape.

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.