Communal Roosts Research Articles

A novel method for estimating avian roost sizes using passive acoustic recordings using deep neural network

Abstract Communal bird roosts serve as information centres and a means of thermoregulation for many species. While some communally roosting species are major pests and cause dis‐amenities, others are of conservation concern. Estimating the population of roosting birds can provide a useful proxy of population size and possibly a more reliable estimate than other sampling techniques. However, estimating these populations is challenging as some roosts are large and often occluded in foliage. Previous acoustic methods such as paired sampling, microphone arrays and use of call rate have been used to estimate bird abundances; however, these are less suited for estimating large roost populations where hundreds of individuals are calling in unison. To address this challenge, we explored using machine learning techniques to estimate a roost population of the Javan myna, Acridotheres javanicus, an invasive species in Singapore. While one may expect to use sound intensity to estimate roost sizes, it is affected by various factors such as distance to the recorder, local propagation conditions (e.g. buildings and trees), weather conditions, and noise from other sources. Here, we used a deep neural network to extract higher order statistics from the sound recordings and use those to help estimate roost sizes. Additionally, we validated our method using automated visual analysis with a dual‐camera setup and manual bird counts. Our estimated bird counts over time using our acoustic model matched the automated visual estimates and manual bird counts at a selected Javan myna roost, thus validating our approach. Our acoustic model estimated close to 400 individual mynas roosting in a single tree. Analyses of additional recordings of Javan myna roosts conducted on two separate occasions and at a different roost location using our acoustic model showed that our roost estimates over time also matched our automated visual estimates well. Practical implication: Our novel approach of estimating communal roost sizes can be achieved robustly using a simple portable acoustic recording system. Our method has multiple applications such as testing the efficacy of avian roost population control measures (e.g. roost tree pruning) and monitoring the populations of threatened bird species that roost communally.

‘Night vision’ cameras as a tool for consistent and repeatable counts at communal overnight roosts for parrots

‘Night vision’ cameras as a tool for consistent and repeatable counts at communal overnight roosts for parrots

Going it alone: observations of solitary roosting in two Australian flying-foxes (Pteropus alecto and P. poliocephalus)

Unlike some flying-fox species that exhibit solitary roosting, all flying-fox species of the Australian mainland are strongly colonial, though some extent of solitary roosting in these species is anecdotally known. This note reports on five accounts of solitary roosting in black flying-foxes (Pteropus alecto) and grey-headed flying-foxes (Pteropus poliocephalus), herein referred to as 'singletons'. These accounts are potentially the first published detailed observations of singletons in Australian flying-foxes outside of visible periods of food shortages. In particular, one account involved solitary roosting recorded over multiple days. The frequency and extent of singletons in Australian flying-foxes remain completely unknown. In terms of deriving population sizes in these species from counts at communal roosts, these individuals represent an addition to those estimates.

Testing the information centre hypothesis in a multilevel society.

In various animal species conspecifics aggregate at sleeping sites. Such aggregations can act as information centres where individuals acquire up-to-date knowledge about their environment. In some species, communal sleeping sites comprise individuals from multiple groups, where each group maintains stable membership over time. We used GPS tracking to simultaneously record group movement in a population of wild vulturine guineafowl (Acryllium vulturinum) to investigate whether communal sleeping sites can facilitate the transfer of information among individuals across distinct groups. These birds live in large and stable groups that move both together and apart, often forming communal roosts containing up to five groups. We first test whether roosts provide the opportunity for individuals to acquire information from members of other groups by examining the spatial organization at roosts. The GPS data reveal that groups intermix, thereby providing an opportunity for individuals to acquire out-group information. We next conduct a field experiment to test whether naïve groups can locate novel food patches when co-roosting with knowledgeable groups. We find that co-roosting substantially increases the chances for the members of a naïve group to discover a patch known to individuals from other groups at the shared roost. Further, we find that the discovery of food patches by naïve groups subsequently shapes their space use and inter-group associations. We also draw on our long-term tracking to provide examples that demonstrate natural cases where communal roosting has preceded large-scale multi-group collective movements that extend into areas beyond the groups' normal ranges. Our findings support the extension of the information centre hypothesis to communal sleeping sites that consist of distinct social groups.

Seed dispersal by the cosmopolitan house sparrow widens the spectrum of unexpected endozoochory by granivore birds.

In the intricate web of plant-animal interactions, granivore birds can play a dual antagonist-mutualist role as seed predators and dispersers. This study delves into the ecological significance of the house sparrow (Passer domesticus) as seed disperser by endozoochory. A sample of individual droppings and faecal pools were collected from a communal roost in central Spain to examine the presence of seeds. Seed viability was determined using the tetrazolium test. Our findings revealed that around 22% of the analysed droppings contained seeds, contradicting the prevalent notion of house sparrow solely as seed predator. Viability tests demonstrated that 53.9% of the defecated seeds were viable, although it varied between plant species, including those from fleshy-fruited common fig and five species of dry-fruited herbs. This study challenges the traditional perspectives on the ecological role of the house sparrow, and glimpses on their contribution to seed dispersal. Understanding the nuanced roles of granivore species like the house sparrow is crucial for developing holistic conservation and management strategies in urban and agricultural landscapes. Future studies are encouraged to unravel the actual role of this cosmopolitan species as disperser of a likely broad spectrum of wild, cultivated and exotic plants.

Pre-Roost Habitats of Western Marsh Harriers (Circus aeruginosus) Wintering in Keoladeo National Park, Bharatpur in Northwestern India

Abstract I recorded information on pre-roost habitats of Western Marsh Harriers (Circus aeruginosus) near communal roost sites located in and around Keoladeo National Park (KNP), India, during three winters from 1997 to 2000. Harriers pre-roosted in different habitats in KNP compared to the surrounding areas (25-km radius). Harriers used trees as their main pre-roosting perches in KNP, but mainly used open fields in the surrounding areas. All nine tree species present near roost sites in KNP were used for pre-roosting. Among tree species used for pre-roosting, harriers preferred sacred fig (Ficus religiosa), a tall tree with a wide canopy. Babul (Vachellia nilotica) trees, though not preferred, were more utilized for pre-roosting, probably because of their abundance. The lack of suitable open areas near the primary roost site in KNP may help explain why harriers used trees as pre-roosting perches there. At other roost sites outside KNP where trees were sparse, harriers mainly pre-roosted on the ground. By perching on trees, harriers may be conspicuous to other arriving harriers, which likely helps the birds gather into a large roosting flock. Such large roosts may benefit harriers by providing protection from predators and a site where birds may obtain information on food resources. Although trees were used as pre-roosting sites at KNP, I did not observe any harriers roosting on trees overnight during the 3-yr study.

Roosting Behaviour and Population of Indian Mynas; & its Correlation with Sunset (light intensity) and the Time of Arrival

Abstract: Communal roosting behaviour is a very fascinating phenomenon that has been observed in numerous avian species wherein many birds stay together during night or daytime. Size of communal roosts may range from 5-6 individuals (e.g. Peafowl, drongo) to over 10,000 birds (e.g. Indian myna, crow, parakeets). Such birds have tendency to flock on their roost-ward journey. In this research, we observed communal roosting behaviour and population of Indian Myna (Acridotheres tristis). We identified one of the communal roosts of Indian Myna in Vaikunth Crematorium, Navi Peth area Pune and counted their population size at the time of their arrival to the roost during sunset on a fortnightly basis for a year by point count method. Census records were taken at 5 minutes interval on the number of flocks and flock sizes of these birds arriving to the communal roost in the evening. Also, recorded the light intensities over those intervals with the help of photometer. Our objective was to find out their population size and changes in the size at this particular roost and also to find out the correlation between the time of sunset (light intensity) and their time of arrival to the roost. This study illustrates the effect of light intensity on their roosting behaviour and analysis of any seasonal and monthly changes observed during the data collection period from 2 consecutive post breeding periods.

Population Size, Non-Breeding Fraction, and Productivity in a Large Urban Population of Burrowing Parrots (Cyanoliseus patagonus)

Psittaciformes are one of the bird orders with the highest number of threatened species and the most marked declining population trends. At present, the lack of information on the population size, reproductive fraction, and productivity of most parrot populations makes it difficult to design effective conservation actions. In this study, we monitored a population of Burrowing Parrots (Cyanoliseus patagonus) breeding in urbanized habitats in the southwest of Buenos Aires province, Argentina. Every December and February from 2018 to 2023, we counted the individuals arriving at a single communal roost, located in the main park of Bahía Blanca city, which gathers all the parrots breeding in 18–22 colonies within a radius of 20 km. Censuses were conducted before (December) and immediately after the incorporation of juveniles into the flocks (February). Breeding pairs were also counted annually in the colonies, and the average annual productivity and the proportion of juveniles were estimated from surveys in pre-roosting and feeding areas in February. The non-breeding fraction approached half of the population with no statistically significant differences among years (range: 37–53%), and the breeding population showed little annual variation, with a minimum of 1363 and a maximum of 1612 breeding pairs. The proportion of juveniles in the flocks and the estimated productivity showed larger variations among breeding seasons. Our results add insight to the scarce information available on the breeding-to-non-breeding-population ratios in parrots, and birds in general, and show key breeding parameters for a species that is thriving well in urban habitats.

Roosting Behavior of Waterbirds at Ruhuna University Premises in Sri Lanka

Communal roosting is a habit of many waterbirds. Although it is a common behavioural pattern, only a few documentary records are available in Sri Lanka. Therefore, the present study aimed to investigate the roosting behaviour of waterbirds at Ruhuna University Premises, Sri Lanka. The study was conducted from November 2020 to February 2021 covering 20 field visits. During field visits, the abundance of waterbirds, their arrival time, intra-species, and inter-species interactions were observed. The responses of roosting waterbirds to human disturbances and occasional visitors to roosting sites were also recorded. The waterbirds roosted on a single Sonneratia caseolaris tree, located at the waterbody closer to the bank. Eleven species of waterbirds were identified, and the Cattle Egrets (Bubulcus ibis) were prominent (93%). The diversity and species abundance gradually increased during the study period. The highest species richness (11) and diversity (H- 0.8832) were recorded in February. The roosting behavior of birds has been altered by human disturbances. The waterbirds co-existed on the roosting site with inter- and intra-species associations while maintaining hierarchical levels (different height levels of the tree). Sixteen species of occasional visitors were recorded, and they left the site with the arrival of regular waterbirds.

Anthropogenic nesting sites and density of Burrowing Parrot (Cyanoliseus patagonus) in northern Argentinian Patagonia

BackgroundThe expansion of human activities and the development of urban centers are among the main driving forces accounting for the transformation and loss of natural environments. At the same time, and especially for some birds, anthropogenic activity provides new habitat resources. This is the case of the Burrowing Parrot (Cyanoliseus patagonus), in and around Bahía Blanca, a city of ca. 335,000 inhabitants in northern Argentinian Patagonia, where urban and rural quarries and constructed ravines on roadsides are where most of its reproductive activity occurs.MethodsIn this study we monitored anthropogenic nesting sites and estimated the number of breeding pairs from 2018 to 2023 through censuses conducted annually in 23 colonies within a radius of up to 20 km from the communal roost located in the city.ResultsMost of the nesting sites (57%), and the breeding pairs (60 to 80%) were in urban environments, and the remaining in rural areas. Ravines along roadsides and quarries represented the substrate that was most frequently used for nesting. Mean nest density was significantly higher in roadside ravines compared to quarries, and, in turn, higher in urban roadsides compared to rural roadsides.ConclusionAnthropogenic substrates appear as key components for the reproduction of the species, with possible effects on its numbers. The ability of the Burrowing Parrot to reproduce on artificial substrates in the urban environment, and especially the rapid colonization of recently opened sites, represents a new perspective for the conservation and management of its populations.

High voltage pylons as roost sites of the Lesser Kestrel (Falco naumanni) during the pre-migratory period in the Southeastern Balkans

Although the Lesser Kestrel (Falco naumanni, Fleischer, 1818) was a widespread species in Bulgaria, there are no detailed surveys on its biology and ecology like the ones implemented in countries still harbouring abundant populations, such as Spain, Italy, Greece, etc. After 2014 when “Green Balkans”, as a breeder, launched the reinforcement of the species and its recovery in Bulgaria now it is possible to study the species using modern technologies and methods like radio and satellite transmitters. The known roost sites (n=12), the subject of our study, are located from 55 to 745 m above the sea level, as the main part of them is situated about 100 - 200 m above the sea level. The most numerous roost site, which is located in the North-eastern Greece, is comprised of about 80 individuals but in the other roost sites the number of the Lesser Kestrel reach 20-40 individuals. During the period 2014 - 2021 the Lesser Kestrels in the largest known colony of the species in Bulgaria, which is located in SPA Sakar, have used different high voltage pylons from the same power lines, alternating pylons depending on the type of agricultural crops in the area. It has been found that the birds express preferences for cereal crops around high voltage pylons, which they use as roost sites. Keywords: communal roost sites, LIFE for Lesser Kestrel, post-breeding dispersal.

Artificially raised roost temperatures lead to larger body sizes in wild bats

Climate warming has major consequences for animal populations, as ambient temperature profoundly influences all organisms' physiology, behavior, or both.1 Body size in many organisms has been found to change with increased ambient temperatures due to influences on metabolism and/or access to resources.2,3,4,5,6 Changes in body size, in turn, can affect the dynamics and persistence of populations.7 Notably, in some species, body size has increased over the last decades in response to warmer temperatures.3,8 This has primarily been attributed to higher food availability,3 but might also result from metabolic savings in warmer environments.9,10 Bechstein's bats (Myotis bechsteinii) grow to larger body sizes in warmer summers,11 which affects their demography as larger females reproduce earlier at the expense of a shorter life expectancy.12,13 However, it remains unclear whether larger body sizes in warmer summers were due to thermoregulatory benefits or due to increased food availability. To disentangle these effects, we artificially heated communal day roosts of wild maternity colonies over four reproductive seasons. We used generalized mixed models to analyze these experimental results along with 25 years of long-term data comprising a total of 741 juveniles. We found that individuals raised in heated roosts grew significantly larger than those raised in unheated conditions. This suggests that metabolic savings in warmer conditions lead to increased body size, potentially resulting in the decoupling of body growth from prey availability. Our study highlights a direct mechanism by which climate change may alter fitness-relevant traits, with potentially dire consequences for population persistence.

Integrated Approach Improves Monitoring of Wintering Red Kites in Central Europe

ABSTRACT The Red Kite (Milvus milvus) is a common raptor in suitable areas of Austria, the Czech Republic, and Slovakia during the nonbreeding season. However, the exact number of birds wintering in these countries remains unknown. Through an integrated approach using satellite telemetry, direct field surveys at nocturnal roosts, and citizen science databases, we estimated the wintering population of Red Kites in this area during the winter of 2020/2021. Based on three surveys taking place at the beginning (28 November 2020), middle (9 January 2021), and end (5 February 2021) of winter, respectively, we counted 566, 558, and 536 Red Kites. The observations from online citizen science databases mostly appeared to correspond with the area of communal roost sites. Using our combined approach, the estimates of wintering Red Kites were determined as minimum numbers in the study area. A comparison of historical data (from the second half of the 20th and beginning of the 21st centuries, respectively) and our own results suggests that the wintering population of Red Kites in the study area has increased substantially. This dynamic could be caused by changes in climatic conditions, a shift of the birds' breeding range in Europe, changes in migration behavior and/or food availability, and/or conservation efforts.

Insight in a behaviour of Central European Red Kites wintering in Italy

Migration and over-wintering strategy of raptors can be influenced by many factors such as sex, age and experience of birds, habitat variability and food availability. Based on the telemetry data of nine Red Kites (Milvus milvus), we observed in situ their wintering grounds in southern Italy during winter 2019/2020. Red Kites originated from Austria and the Czech Republic arrived to their wintering grounds from 10th September to 5th December 2019 and departed between 9th February and 7th May 2020. Tagged Kites stayed in their wintering grounds from 73 to 207 days (mean 135±46 SD). These birds occupied home range and the core area from 39 to 9,351 km2 (KDE 95%; mean 3,591±3,460 SD) and 6 to 1,998 km2 (KDE 50%; mean 566±483 SD), respectively. We observed a total of 242 Red Kites at nine communal roost sites (min. = 14, max. = 45 individuals) used by tagged Red Kites. Roosts were situated at an average altitude of 630 m above sea level. Agricultural countryside with small or large fields, small or large forest patches, meadows, pastures, sheep or cattle farms surrounded areas around communal roosts of Red Kites.

Social situations differ in their contribution to population-level social structure in griffon vultures.

Social relationships among animals emerge from interactions in multiple ecological and social situations. However, we seldom ask how each situation contributes to the global structure of a population, and whether different situations contribute different information about social relationships and the position of individuals within the social fabric. Griffon vultures (Gyps fulvus) interact socially in multiple situations, including communal roosting, joint flights, and co-feeding. These social interactions can influence population-level outcomes, such as disease transmission and information sharing that determine survival and response to changes. We examined the unique contribution of each social and ecological situation to the social structure of the population and individuals' positions within the overall social network using high-resolution GPS tracking. We found that the number of individuals each vulture interacted with (degree) was best predicted by diurnal interactions-both during flights and on the ground (such as when feeding). However, the strength of social bonds, that is, the number of interactions an individual had (strength), was best predicted by interactions on the ground-both during the day (e.g., while feeding) and at night (e.g., while roosting) but not by interactions while flying. Thus, social situations differ in their impact on the relationships that individuals form. By incorporating the ecological situations in which social interactions occur we gain a more complete view of how social relationships are formed and which situations are important for different types of interactions.

A Kist o’ Whistles

A Kist o’ Whistles

Effects of the environmental conditions and seasonality on a population survey of the Andean condor Vultur gryphus in the tropical Andes.

Among the New World vultures, the Andean condor is considered one of the most culturally and ecologically important species. However, their populations are declining over their entire distributional range. In response, conservation strategies have been implemented in many countries to reverse the increasing extinction risk of this species. The initiatives rely on extensive population surveys to gather basic information necessary to implement policies and to intervene efficiently. Still, there is a need to standardize the surveys based on seasonality and suitable environmental conditions throughout the species distribution. Here, we provide the first assessment of how daily temperature, rainfall, and seasonality influence surveys of Andean condors on a communal roost in the central Peruvian Andes. Using an autoregressive generalized linear model, we associated environmental variables with visual surveys of adult and young condors at three different times of the day and three times a week between June 2014 and March 2015. We found that both adults and young Andean condors showed a threefold reduction in the use of the communal roost after the beginning of the rainy season. Colder and drier days (dry season) are preferable for surveying, as we expect the total number of condors using communal roosts to reduce under rainy (rainfall = -0.53 ± 0.16) and warmer days (temperature = -0.04 ± 0.02) days. Therefore, the significant variation in the use of roosts across seasons and hours should be carefully accounted for in national surveys, at the risk of undermining the full potential of the communal roost surveys. Moreover, we also found a strong bias towards immatures (about 76%) in the adult:immature ratio and a remarkable absence of Andean condors during the wet season. These results suggest that the species might be using other unknown communal roosts hierarchically. Such results provide key information for selecting priority areas for conservation and selecting the best time to survey this species in the tropical Andes. Finally, it may open a fruitful avenue for further research on the protection of the Andean condor.

Using computer vision, image analysis and UAVs for the automatic recognition and counting of common cranes (Grus grus)

Long-term monitoring of wildlife numbers traditionally uses observers, which are frequently inefficient and inaccurate due to their variable experience/training, are costly and difficult to sustain over time. Furthermore, there are other inhibiting factors for wildlife counting, such as: inhabiting inaccessible areas, fear of humans, and nocturnal behavior. There is a need to develop new technologies that will automatically identify and count wild animals in order to determine the appropriate management protocol. In this study, an advanced and accurate method for automatically calculating the number of cranes (Grus grus), using thermal cameras at night and visible light (RGB) cameras during the day onboard unmanned aerial vehicles (UAVs), based on image analysis and computer vision, was developed. The cranes congregate at night in a large communal roost, making it possible to count the birds while they are relatively static and all together. Each bird was counted individually by creating a standardized tool to determine population numbers for management, using image analysis and automatic processing. A dedicated algorithm was developed that aimed to identify the cranes based on their spectral characteristics (typical temperature, shape, size) and to effectively separate the cranes from the typical background. The automatic segmentation and counting of roosting common cranes using UAV nighttime thermal images had an Overall Accuracy (OA) of 91.47%, User's Accuracy (UA) of 99.68%, and Producer's Accuracy (PA) of 91.74%. The computer vision and machine learning algorithm based on the YOLO v3 platform of daytime RGB UAV images of common cranes at the feeding station yielded an overall loss accuracy level of 2.25%, with a mean square error of 1.87, OA of 94.51%, UA of 99.91%, PA of 94.59%. These results are highly encouraging, and although the algorithms were developed for the purpose of counting cranes, they could be adapted for other counting purposes for wildlife management.

Discriminant Criteria for Field Sexing in the Eurasian Tree Sparrow by Combining Body Size and Plumage Features

The Eurasian tree sparrow (Passer montanus) is a monomorphic passerine, for which it is impossible to differentiate between males and females based on external characteristics. Being a species frequently captured for ringing, having a reliable method to determine sex from conventional biometric measurements would facilitate its study and be very useful for the correct management and conservation of this declining species. In the present study, we used biometric measurements recorded in 66 individuals captured with mist nets in communal roosts in northern Spain during the winter and sexed them using molecular techniques. We conducted a discriminant function analysis (DFA) to derive equations that allowed us to determine the sex of the specimens from some of the measurements recorded in the field. Significant differences were found between males and females in wing length, third primary length, badge width and height and body weight. The DFA provided two functions that correctly classified the sex of 94.7% of the individuals using wing length and badge width, and 98.2% if weight was added to the analysis. Our results allow sexing from measurements that can be easily recorded in the field with the tools commonly used in banding sessions and without the need for additional training. Considerations of Bergmann’s and Allen’s rules on body size and the use of DFA in different populations are discussed.

Social network analysis reveals context-dependent kin relationships in wild sulphur-crested cockatoos Cacatua galerita.

A preference to associate with kin facilitates inclusive fitness benefits, and increased tolerance or cooperation between kin may be an added benefit of group living. Many species exhibit preferred associations with kin; however, it is often hard to disentangle active preferences from passive overlap, for example caused by limited dispersal or inheritance of social position. Many parrots exhibit social systems consisting of pair-bonded individuals foraging in variably sized fission-fusion flocks within larger communal roosts of hundreds of individuals. Previous work has shown that, despite these fission-fusion dynamics, individuals can exhibit long-term preferred foraging associations outside their pair bonds. Yet the underlying drivers of these social preferences remain largely unknown. In this study, we use a network approach to examine the influence of kinship on social associations and interactions in wild, communally roosting sulphur-crested cockatoos, Cacatua galerita. We recorded roost co-membership, social associations and interactions in 561 individually marked birds across three neighbouring roosts. We then collected genetic samples from 205 cockatoos, and conducted a relationship analysis to construct a kinship network. Finally, we tested correlations between kinship and four social networks: association, affiliative, low-intensity aggression and high-intensity aggression. Our result showed that while roosting groups were clearly defined, they showed little genetic differentiation or kin structuring. Between roost movement was high, with juveniles, especially females, repeatedly moving between roosts. Both within roosting communities, and when visiting different roosts, individuals preferentially associated with kin. Supporting this, individuals were also more likely to allopreen kin. However, contrary to expectation, individuals preferred to direct aggression towards kin, with this effect only observed when individuals shared roost membership. By measuring social networks within and between large roosting groups, we could remove potential effects of passive spatial overlap on kin structuring. Our study reveals that sulphur-crested cockatoos actively prefer to associate with kin, both within and between roosting groups. By examining this across different interaction types, we further demonstrate that sulphur-crested cockatoos exhibit behavioural and context-dependent interaction rules towards kin. Our results help reveal the drivers of social association in this species, while adding to the evidence for social complexity in parrots.