Habitat correlates of pulse parameters in the highly specialised acoustic system of Chiroptera

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
High Duty Cycle echolocating bats use high frequency echolocation pulses that have limited range but are clutter resistant i.e. can detect targets in dense clutter (the number of echoes produced other than those from the target of interest). A specialised echolocation pulse design (consisting of a constant frequency and frequency modulated components) facilitates foraging for prey in habitats characterised by dense vegetation or clutter. The constant frequency component, along with an acoustic fovea and Doppler Shift Compensation, allows them to distinguish small moving targets from stationary background objects. The frequency modulated components are thought to be used for depth resolution (i.e. ranging acuity). In contrast to their clutter specialist status, these species are found in a variety of biomes including open desert. A negative correlation between level of environmental clutter and frequency has been established for some species. The Foraging Habitat Hypothesis (FHH) proposes that the evolution of echolocation frequency is linked with changes in habitat clutter. In High Duty Cycle bats, the FHH predicts areas of low clutter should select lower frequency pulses because they are less susceptible to atmospheric attenuation making them more suited to long distance prey detection. Previous research has therefore focused on the methods by which bats vary the frequency of their pulses to achieve optimal propagation distances. However, the source levels of these signal emissions are also under control of the bat and must play a major role in signal propagation and therefore in detection of prey. My study tested the FHH by combining both an observational and experimental approach to determine how habitat clutter influences echolocation pulse divergence in High Duty Cycle bats. My focal species was Rhinolophus capensis, which has previously been shown to use different pulse frequencies apparently associated with differences in habitat structure. I focused on two populations, R. capensis in the fynbos (pulse frequency: 84 kHz) and R. capensis in the desert (pulse frequency: 74 kHz). Bats were recorded using a multiple microphone array in their natural habitat and in a flight room experiment where they were exposed to both a cluttered (simulating the fynbos biome) and open (simulating the desert biome) flight room. The experiment determined whether observed differences were a result of possible selection (as proposed by the FHH) or behavioural flexibility. A congeneric species, iii R. damarensis, was used as a control and additional test of the FHH because it inhabits the same desert biome as R. capensis but echolocates at a higher frequency (equivalent to the frequency used by R. capensis in the fynbos). In accordance with this hypothesis, I also tested if there were differences in the frequency modulated components of R. capensis pulses between biomes and whether these differences were due to possible selection for optimal ranging acuity in response to the degree of clutter in each biome. My results suggest the use of lower frequency echolocation pulses in R. capensis in the desert could have evolved for increased detection distance (as proposed by the FHH) but that lower frequencies may not be a prerequisite for successful foraging in open biomes. In R. capensis the greatest differences in prey detection between biomes was a product of both frequency and source level. However, higher source levels used by R. capensis in the desert had a greater contribution to observed differences in detection distances both within (emergence versus foraging area, cluttered versus open flight room) and between biomes (desert versus fynbos) than frequency. In addition, on average R. damarensis did not compensate for higher frequencies with higher source levels resulting in lower average detection distances than R. capensis in the desert. However, a few measurements of source levels for R. damarensis were the highest recorded and resulted in the largest prey detection distances recorded in my study. These findings support recent findings that suggest that SLs are energetically costly. In both biomes, bats used lower source levels when exposed to a cluttered versus open flight room. In the desert biome, bats actively lowered their source levels (compared to the source levels they use in the field) when exposed to a level of clutter that they do not experience naturally (cluttered flight room treatment). Unlike source levels, frequency (of the constant frequency component) was conserved during the flight room treatments. Frequency differences between R. capensis in the different biomes can therefore be attributed to possible selection rather than behavioural flexibility. The conservation of frequency prompted bats to vary their source levels to achieve appropriate detection distances when exposed to different environmental stimuli. Source level flexibility may therefore be the key to the capability of specialist clutter foragers to successfully hunt and survive in harsh open environments. To the same extent that source levels facilitate foraging in open environments, the frequency modulated components of High Duty Cycle bat pulses may facilitate the orientation and foraging of these bats in cluttered biomes. In accordance with the FHH, a strong correlation was found between the frequency (i.e. number of occurrences)/bandwidth of these components and the level of environmental clutter both within iv (between the two treatments of the experiment) and between biomes. The findings in my study highlight the importance of environmental pressures, such as clutter, in shaping the echolocation pulse parameters of bats.