Estimating population density of insectivorous bats based on stationary acoustic detectors: A case study

A simple and fast method for estimating bat roost locations

Bats play a pivotal role in pest control, pollination and seed dispersal. Despite their ecological significance, locating bat roosts remains a challenging task for ecologists. Traditional field surveys are time-consuming, expensive and may disturb sensitive bat populations. In this article, we combine data from static audio detectors with a bat movement model to facilitate the detection of bat roosts. Crucially, our technique not only provides a point prediction for the most likely location of a bat roost, but because of the algorithm's speed, it can be applied over an entire landscape, resulting in a likelihood map, which provides optimal searching regions. To illustrate the success of the algorithm and highlight limitations, we apply our technique to greater horseshoe bat (Rhinolophus ferrumequinum) acoustic data acquired from six surveys from four different UK locations and over six different times in the year. Furthermore, we investigate what happens to the accuracy of our predictions in the case that the roost is not contained within the area spanned by the detectors. This innovative approach to searching rural environments holds the potential to greatly reduce the labour required for roost finding, and, hence, enhance the conservation efforts of bat populations and their habitats.

Acoustic monitoring yields informative bat population density estimates

Bat population estimates are typically made during winter, although this is only feasible for bats that aggregate in hibernacula. While it is essential to measure summer bat population sizes for management, we lack a reliable method. Acoustic surveys should be less expensive and more efficient than capture surveys, and acoustic activity data are already used as indices of population size. Although we currently cannot differentiate individual bats by their calls, we can enter call counts, information on signal and detection angles, and weather data into generalized random encounter models to estimate bat density. We assessed the utility of generalized random encounter models for estimating Indiana bat (Myotis sodalis) population density with acoustic data collected at 51 total sites in six conservation areas in northeast Missouri, 2019-2021. We tested the effects of year, volancy period, conservation area, and their interactions on estimated density. Volancy period was the best predictor, with average predicted density increasing 60% from pre-volancy (46 bats/km2) to post-volancy (74 bats/km2); however, the magnitude of the effect differed by conservation area. We showed that passive acoustic surveys yield informative density estimates that are responsive to temporal changes in bat population size, which suggests this method may be useful for long-term monitoring. However, we need more information to choose the most appropriate values for the density estimation formula. Future work to refine this approach should include assessments of bat behavior and detection parameters and testing the method's efficacy in areas where population sizes are known.

Estimating density of forest bats and their long-term trends in a climate refuge

For many species, estimating density is challenging, but it is important for conservation planning and understanding the functional role of species. Bats play key ecological roles, yet little is known about their free-ranging density. We used a long-term banding study of four species caught in an extensively forested climate refuge and spatial capture-recapture models (SCR) to estimate density and its change over time. Between 1999 and 2020, there were 3671 captures of four bat species, which were all edge-space foragers. Recaptures represented 16% (n = 587) of all captures, of which 89 were between-trap-cluster movements. Closed spatial mark-recapture models estimated plausible densities that varied with elevation. Preferred elevations differed between species, with density averaging 0.63 ha^-1 for Vespadelus darlingtoni (high elevation), 0.43 ha^-1 for V. pumilus (low elevation), 0.19 ha^-1 for Chalinolobus morio (high elevation), and 0.08 ha^-1 for V. regulus (high elevation). Overall, densities were higher than most previous published estimates for bats. Forest disturbance history (past timber harvesting) had no detectable effect on density. Density also varied substantially across years, and although annual maximum temperature and rainfall were not supported in models, some time periods showed an apparent relationship between density and annual rainfall (+ve) and/or annual maximum temperature (-ve). The most notable change was an increase in the density of V. pumilus after 2013, which tracked an increase in annual temperature at the site, reflecting a warming climate. Bat densities in forests outside of climate refugia are likely to be more sensitive to climate change, but more studies are needed in different habitats and continents and outside climate refugia to place the densities we estimated into a broader context.

Fruit bats adjust their foraging strategies to urban environments to diversify their diet

Background

Urbanization is one of the most influential processes on our globe, putting a great number of species under threat. Some species learn to cope with urbanization, and a few even benefit from it, but we are only starting to understand how they do so. In this study, we GPS tracked Egyptian fruit bats from urban and rural populations to compare their movement and foraging in urban and rural environments. Because fruit trees are distributed differently in these two environments, with a higher diversity in urban environments, we hypothesized that foraging strategies will differ too.

Results

When foraging in urban environments, bats were much more exploratory than when foraging in rural environments, visiting more sites per hour and switching foraging sites more often on consecutive nights. By doing so, bats foraging in settlements diversified their diet in comparison to rural bats, as was also evident from their choice to often switch fruit species. Interestingly, the location of the roost did not dictate the foraging grounds, and we found that many bats choose to roost in the countryside but nightly commute to and forage in urban environments.

Conclusions

Bats are unique among small mammals in their ability to move far rapidly. Our study is an excellent example of how animals adjust to environmental changes, and it shows how such mobile mammals might exploit the new urban fragmented environment that is taking over our landscape.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12915-021-01060-x.