Stronger together: Combining automated classifiers with manual post-validation optimizes the workload vs reliability trade-off of species identification in bat acoustic surveys

Some bats are here: Reducing the Wallacean shortfall of bats in the amazon

The Amazon rainforest has approximately 23% of its sampled area dedicated to bats, making it one of the least sampled and most diverse regions for bats in Brazil. The lack of sampling results in a lack of knowledge regarding the accurate geographical distribution of bat species. This lack is referred to as the Wallacean shortfall, which should be addressed with primary data obtained from in situ collections. However, the use of Species Distribution Models (SDMs) can help alleviate this gap. The states of Pará and Acre are located in the Brazilian Amazon. So, our objective is to decrease the Wallacean shortfall concerning Amazonian bat species. To achieve this, we provide (i) a list of bat species sampled in the states of Pará and Acre in the last 5 years (2017 to 2022); (ii) the potential distribution of species considered as new occurrences for the region; and (iii) the potential distribution of species classified as Data Deficient (DD) and Near Threatened (NT) according to the IUCN classification. With 96 nights of collection and 129,600 m2h of mist netting, we obtained 75 bat species, with an estimated total of 94.78 species. Additionally, 21 species were considered as range extensions. The Brazilian Amazon region has a vast geographic expanse and few established research centers, resulting in a limited sampling of bats and other biological groups. Furthermore, we draw attention to the significant number of bat species with expanded geographical distributions, with 21 out of the 75 sampled species. This should be a reminder that primary biogeographic data is still necessary for the neotropical region.

Unexpected bat community changes along an urban-rural gradient in the Berlin-Brandenburg metropolitan area

Urbanization gradients are increasingly used in ecological studies to discover responses of species communities to different intensities of human-induced habitat transformation. Here, we investigated patterns of bat communities against the background of different urbanization levels using a priori defined urbanization categories based on distance classes (5 km intervals) along a linear transect from the urban core of the city of Berlin westwards into the rural outskirts of the state of Brandenburg. Using linear-mixed effects models, we found that "distance class", as a proxy for urbanization level, is a meaningful and suitable predictor of bat species richness and diversity. We observed an unexpectedly sudden increase in bat species richness and diversity and changes in species-specific activity levels relatively close to the urban center at the transition between urban and peri-urban areas. This change suggests a relevant influence of the peri-urban areas as a "buffer zone" for specific bat species not able to adapt to the heavily modified inner core of the metropolitan area. Although we could demonstrate that anthropogenic noise and artificial light have the potential to predict the variability of bat species activity along the urban-rural gradient, the actual influence on observed shifts in the bat community needs further research.

Bats at an Altitude above 2000 m on Pirin Mountain, Bulgaria

The study describes a pilot survey on bats in the highest areas of Pirin Mountain. The methods included examining subfossil bone remains, mist-netting, and recording echolocation calls. The study was conducted in August 2002 and 2013 and from 2019 to 2020. While in general, bat diversity tends to decrease with increasing altitude due to harsher environmental conditions, the present study, despite a short period, reveals high diversity. Twenty species, more than half of the Bulgarian bat fauna, were detected. The recording and analysis of vocal signatures proved to be the best way to inventory bat diversity. At least 13 species were detected by this method. Vespertilio murinus and Tadarida teniotis together make up more than 60% of all reliably determined echolocation sequences. Significant activity was found for Myotis myotis/blythii, Plecotus auritus, Eptesicus serotinus, and E. nilssonii. The registration of the latter species is of considerable faunistic interest. It was previously only known from a single specimen at one location in the country. The sex and age structure of the bat assemblage suggests that it is likely a swarming assemblage. The area is the highest swarming location in Europe. The results provide valuable information on bat ecology and behaviour, which can be used to inform management and protection efforts.

Systematic review of machine learning methods applied to ecoacoustics and soundscape monitoring

Soundscape ecology is a promising area that studies landscape patterns based on their acoustic composition. It focuses on the distribution of biotic and abiotic sounds at different frequencies of the landscape acoustic attribute and the relationship of said sounds with ecosystem health metrics and indicators (e.g., species richness, acoustic biodiversity, vectors of structural change, gradients of vegetation cover, landscape connectivity, and temporal and spatial characteristics). To conduct such studies, researchers analyze recordings from Acoustic Recording Units (ARUs). The increasing use of ARUs and their capacity to record hours of audio for months at a time have created a need for automatic processing methods to reduce time consumption, correlate variables implicit in the recordings, extract features, and characterize sound patterns related to landscape attributes. Consequently, traditional machine learning methods have been commonly used to process data on different characteristics of soundscapes, mainly the presence-absence of species. In addition, it has been employed for call segmentation, species identification, and sound source clustering. However, some authors highlight the importance of the new approaches that use unsupervised deep learning methods to improve the results and diversify the assessed attributes. In this paper, we present a systematic review of machine learning methods used in the field of ecoacoustics for data processing. It includes recent trends, such as semi-supervised and unsupervised deep learning methods. Moreover, it maintains the format found in the reviewed papers. First, we describe the ARUs employed in the papers analyzed, their configuration, and the study sites where the datasets were collected. Then, we provide an ecological justification that relates acoustic monitoring to landscape features. Subsequently, we explain the machine learning methods followed to assess various landscape attributes. The results show a trend towards label-free methods that can process the large volumes of data gathered in recent years. Finally, we discuss the need to adopt methods with a machine learning approach in other biological dimensions of landscapes.

What drives bat activity at field boundaries?

Field boundaries are important habitat for bats in agricultural landscapes, serving as commuting and foraging areas for many species. The goal of our study was to better understand the drivers of bat activity in agricultural landscapes to inform conservation policy and make specific recommendations for habitat management. We placed sixty-four full spectrum bat detectors at random recording locations, weekly, along field boundaries in North Somerset between July and October 2020. We used an automated classifier to analyse recordings and performed error rate modelling to account for and remove the majority of error in automated classifications. We used generalised additive models to explore bat response to recording location metrics, controlling for spatial-autocorrelation and temporal differences in sampling. We validated our models with k-fold cross-validation and explored the utility of our models for predicting bat activity at new sites. We found that field boundary characteristics better describe bat activity than adjacent field characteristics or measures of local landscape for the majority of species studied. Bat activity was higher along tall, wide, vegetated field boundaries containing trees; there was lower activity at arable recording locations. Still, bat activity was highly variable and predictive error was high. We found a large spatial effect driving activity patterns, meaning models are not able to predict activity beyond the extent of the study area. We recommend management strategies that give incentives to farmers for replacing fences with hedgerows, planting hedgerow trees, and maintaining tall and outgrown field boundaries.

Edge effects and vertical stratification of aerial insectivorous bats across the interface of primary-secondary Amazonian rainforest

Edge effects, abiotic and biotic changes associated with habitat boundaries, are key drivers of community change in fragmented landscapes. Their influence is heavily modulated by matrix composition. With over half of the world’s tropical forests predicted to become forest edge by the end of the century, it is paramount that conservationists gain a better understanding of how tropical biota is impacted by edge gradients. Bats comprise a large fraction of tropical mammalian fauna and are demonstrably sensitive to habitat modification. Yet, knowledge about how bat assemblages are affected by edge effects remains scarce. Capitalizing on a whole-ecosystem manipulation in the Central Amazon, the aims of this study were to i) assess the consequences of edge effects for twelve aerial insectivorous bat species across the interface of primary and secondary forest, and ii) investigate if the activity levels of these species differed between the understory and canopy and if they were modulated by distance from the edge. Acoustic surveys were conducted along four 2-km transects, each traversing equal parts of primary and ca. 30-year-old secondary forest. Five models were used to assess the changes in the relative activity of forest specialists (three species), flexible forest foragers (three species), and edge foragers (six species). Modelling results revealed limited evidence of edge effects, except for forest specialists in the understory. No significant differences in activity were found between the secondary or primary forest but almost all species exhibited pronounced vertical stratification. Previously defined bat guilds appear to hold here as our study highlights that forest bats are more edge-sensitive than edge foraging bats. The absence of pronounced edge effects and the comparable activity levels between primary and old secondary forests indicates that old secondary forest can help ameliorate the consequences of fragmentation on tropical aerial insectivorous bats.

Three-dimensional stratification pattern in an old-growth lowland forest: How does height in canopy and season influence temperate bat activity?

The study of animal-habitat interactions is of primary importance for the formulation of conservation recommendations. Flying, gliding, and climbing animals have the ability to exploit their habitat in a three-dimensional way, and the vertical canopy structure in forests plays an essential role for habitat suitability. Forest bats as flying mammals may seasonally shift their microhabitat use due to differing energy demands or changing prey availability, but the patterns are not well understood. We investigated three-dimensional and seasonal habitat use by insectivorous bats in a temperate lowland old-growth forest, the Belovezhskaya Pushcha in Belarus. We acoustically sampled broadleaved and mixed coniferous plots in the forest interior and in gaps in three heights during two reproductive periods (pregnancy/lactation vs. postlactation). In canopy gaps, vertical stratification in bat activity was less pronounced than in the forest interior. Vertical activity patterns differed among species. The upper canopy levels were important foraging habitats for the open-space forager guild and for some edge-space foragers like the Barbastelle bat Barbastella barbastellus and the soprano pipistrelle Pipistrellus pygmaeus. Myotis species had highest activity levels near the ground in forest gaps. Moreover, we found species-dependent seasonal microhabitat shifts. Generally, all species and species groups considered except Myotis species showed higher activity levels during postlactation. Myotis species tended toward higher activity in the forest interior during postlactation. P. pygmaeus switched from high activity levels in the upper canopy during pregnancy and lactation to high activity levels near the ground during postlactation. We conclude that a full comprehension of forest bat habitat use is only possible when height in canopy and seasonal patterns are considered.

Are bat mist nets ideal for capturing bats? From ultrathin to bird nets, a field test

The use of mist nets is the most widespread technique to capture bats; however, no study has compared if the type of ground-level mist net used during sampling affects bat captures. We sampled bats using three different types of mist nets that varied in mesh (16, 18, and 20 mm) and denier/ply (45/1 and 75/2) sizes over 76 half-night surveys. We used 17–20 mist nets and checked them at intervals of 15–20 min. Capture rate for echolocating bats was higher in the two mist nets with the biggest denier/ply and smaller mesh sizes. “Ultrathin bat mist nets” showed the lowest capture rates (1.5 times less than “Regular bat mist nets”), whereas “Bird mist nets” had capture rates only 1.2 times smaller than “Regular bat mist nets.” Our results showed that “Bird mist nets” can sample echolocating bats almost as well as “Regular bat mist nets,” and that thinner mist nets may not be the best solution to capture bats that echolocate at high frequencies in this type of surveys. We highlight the importance of considering the efficiency, durability, and longevity of mist nets when choosing the ideal mist-net type for a bat survey.

Optimizing bat bioacoustic surveys in human-modified Neotropical landscapes

During the last decades, the use of bioacoustics as a non-invasive and cost-effective sampling method has greatly increased worldwide. For bats, acoustic surveys have long been known to complement traditional mist-netting, however, appropriate protocol guidelines are still lacking for tropical regions. Establishing the minimum sampling effort needed to detect ecological changes in bat assemblages (e.g., activity, composition, and richness) is crucial in view of workload and project cost constraints, and because detecting such changes must be reliable enough to support effective conservation management. Using one of the most comprehensive tropical bat acoustic data sets, collected in the Amazon, we assessed the minimum survey effort required to accurately assess the completeness of assemblage inventories and habitat selection in fragmented forest landscapes for aerial insectivorous bats. We evaluated a combination of 20 different temporal sampling schemes, which differed regarding number of hours per night, number of nights per site, and sampling only during the wet or dry season, or both. This was assessed under two different landscape scenarios: in primary forest fragments embedded in a matrix of secondary forest and in the same forest fragments, but after they had been re-isolated through clearing of the secondary forest. We found that the sampling effort required to achieve 90% inventory completeness varied considerably depending on the research aim and the landscape scenario evaluated, averaging ~80 and 10 nights before and after fragment re-isolation, respectively. Recording for more than 4 h per night did not result in a substantial reduction in the required number of sampling nights. Regarding the effects of habitat selection, except for assemblage composition, bat responses in terms of richness, diversity, and activity were similar across all sampling schemes after fragment re-isolation. However, before re-isolation, a minimum of four to six sampling hours per night after dusk and three to five nights of sampling per site were needed to detect significant effects that could otherwise go unnoticed. Based on our results, we propose guidelines that will aid to optimize sampling protocols for bat acoustic surveys in the Neotropics.

Collision risk of bats with small wind turbines: Worst-case scenarios near roosts, commuting and hunting structures

Small wind turbines (SWTs) have become increasingly common within the last decade, but their impact on wildlife, especially bats, is largely unknown. We conducted an operational experiment by sequentially placing a mobile SWT with five different operational modes at six sites of high bat activity, including roosts, commuting structures, and highly frequented hunting areas. Bat flight trajectories around the SWT were documented at each site during five consecutive nights using a specifically designed high-spatial-resolution 3D camera. The recordings showed high bat activity levels close to the SWT (7,065 flight trajectories within a 10-m radius). The minimum distance to the rotor of each trajectory varied between 0 and 18 m, with a mean of 4.6 m across all sites. Linear mixed models created to account for site differences showed that, compared to a reference pole without a SWT, bats flew 0.4 m closer to the rotor (95% CI 0.3–0.6 m) if it was out of operation and 0.3 m closer (95% CI 0.1–0.4 m) if it was moving slowly. Exploratory behavior was frequently observed, with many bats deviating from their original flight trajectory to approach the rotor. Among 7,850 documented trajectories, 176 crossed the rotor, including 65 while it was in motion. The collision of one P. pygmaeus individual occurred during the experiment. These results demonstrate that, despite the generally strong ability of bats to evade moving rotor blades, bat casualties at SWTs placed at sites of high bat activity can reach or exceed the current threshold levels set for large wind turbines. As SWTs provide less energy than large turbines, their negative impact on bats should be minimized by avoidance measures such as a bat-friendly site selection or curtailment algorithms.

Time of night and moonlight structure vertical space use by insectivorous bats in a Neotropical rainforest: an acoustic monitoring study

Background

Previous research has shown diverse vertical space use by various taxa, highlighting the importance of forest vertical structure. Yet, we know little about vertical space use of tropical forests, and we often fail to explore how this three-dimensional space use changes over time.

Methods

Here we use canopy tower systems in French Guiana and passive acoustic monitoring to measure Neotropical bat activity above and below the forest canopy throughout nine nights. We use a Bayesian generalized linear mixed effect model and kernel density estimates to demonstrate patterns in space-use over time.

Results

We found that different bats use both canopy and understory space differently and that these patterns change throughout the night. Overall, bats were more active above the canopy (including Cormura brevirostris, Molossus molossus, Peropteryx kappleri and Peropteryx macrotis), but multiple species or acoustic complexes (when species identification was impossible) were more active in the understory (such as Centronycteris maximiliani, Myotis riparius, Pteronotus alitonus and Pteronotus rubiginosus). We also found that most bats showed temporally-changing preferences in hourly activity. Some species were less active (e.g., P. kappleri and P. macrotis), whereas others were more active (Pteronotus gymnonotus, C. brevirostris, and M. molossus) on nights with higher moon illuminance.

Discussion

Here we show that Neotropical bats use habitat above the forest canopy and within the forest understory differently throughout the night. While bats generally were more active above the forest canopy, we show that individual groups of bats use space differently over the course of a night, and some prefer the understory. This work highlights the need to consider diel cycles in studies of space use, as animals use different habitats during different periods of the day.

Bats actively prey on mosquitoes and other deleterious insects in rice paddies: Potential impact on human health and agriculture

BACKGROUND

The fact that bats suppress agricultural pests has been measured for some particular dyads of predator and prey species in both economic and food security terms. The recent emergence of new molecular techniques allows for more precise screenings of bat's diet than the traditional visual identification systems and provides further evidence that bats consume an ample array of agricultural pest species. The main focus of the regulatory services that bats provide in agroecosystems has been on crop pests that cause yield losses. Rice paddies constitute a particular agronomic system with specific challenges, not only related to crop productivity but also to human health. Dipteran density in such ecosystems poses a serious threat to human wellbeing and hinders crop production. Mosquitoes cause direct harm to human populations, transmitting a number of infectious diseases. Non-biting midges (Chironomidae) can consume and weaken rice seedlings and can cause major yield losses.

RESULTS

Mosquito populations and bat activity were assessed in rice paddies of Montgrí, Medes i Baix Ter Natural Park (NE Iberian Peninsula). Molecular analyses of bats faeces (6-weekly samples of 15 faeces each between mid-August and September) proved the presence of both mosquitoes and nonbiting midges in all diet samples. Furthermore, bat activity at the sampling locations was related to adult mosquito density.

CONCLUSION

Our results suggest that bats actively exploit the emergence of adult mosquitoes and further prove that they prey on mosquitoes, nonbiting midges and other deleterious insects. Promoting the presence of bats next to human settlements in such agroecosystems may constitute a biological control system with direct impact on both human health and crop yield. © 2020 Society of Chemical Industry.