Murcielagos

Acta Chiropterologica, 6(2): 347–363, 2004
PL ISSN 1508-1109 © Museum and Institute of Zoology PAS

Data, sample sizes and statistics affect the recognition of species of bats by their echolocation calls
STEFANIA BISCARDI1, JAZMINE ORPRECIO1, M. BROCK FENTON1, 4, ASAF TSOAR2, and JOHN M. RATCLIFFE3
1Department of Biology, York University, North York, Ontario M3J 1P3, Canada Department ofEvolution, Systematics and Ecology, The Hebrew University of Jerusalem, Jerusalem 91904, Israel 3 Department of Zoology, University of Toronto at Mississauga, Mississauga, Ontario L5L 1C6, Canada 4 Corresponding author: E-mail: bfenton@uwo.ca 2

Identification of bat species based on analysis of echolocation calls can be affected by the way data are manipulated, the diversity of species, and callvariability. We document the effects of sample sizes and a priori assignment of calls by species on the outcome of discriminant function analysis (DFA) and multinomial logistic regression (MLR) of features of echolocation calls, and determine which features of calls are most useful for identification. We used recorded echolocation calls of eight species readily distinguishable by call features,including molossids, emballonurids and a moormopid recorded at sites in Belize, Brazil, and Mexico. On individual calls, we measured four features: frequency with most energy, highest and lowest frequencies and call durations obtained from sequences consisting of 10 calls. Cluster analysis and multiple analyses of variance indicated significant differences between the calls of different species.Outcomes of DFA and MLR were affected by both sample sizes (numbers of calls, numbers of sequences) and the subjective approach that researchers take to their data (i.e., categorizing calls or sequences of calls by species). Levels of variation in calls of some species in our sample often precluded the use of single calls in making call-based identifications. Accurate documentation of variability inecholocation behavior of sympatric bats is a prerequisite for an effective sound-based bat survey. Key words: echolocation calls, emballonurids, molossids, mormoopids, variation, cluster analysis, discriminant function analysis, multiple logistic regression

INTRODUCTION Recognizing species of echolocating bats by their calls has been valuable in assessing the distribution of, and habitat use by,bats not prone to capture (e.g., Kalko, 1997; Ochoa et al., 2000). Many authors (e.g., Fenton and Bell, 1981; Ahlén, 1990; O’Farrell and Miller, 1997, 1999; Russo and Jones, 2002; Rydell et al., 2002) have demonstrated that it is possible

to distinguish among species of bats by their echolocation calls, reflecting diagnostic patterns of frequency change over time and species-specificfrequencies in calls. The feasibility of detecting and identifying a bat by its echolocation calls is a direct function of call intensity. Species producing low intensity echolocation calls (the ‘whispering bats’ of Griffin, 1958) are more difficult to sample than bats using high intensity echolocation calls. Among species producing

348

S. Biscardi, J. Orprecio, M. B. Fenton, A. Tsoar, and J. M.Ratcliffe

high intensity echolocation calls, some call features appear to be particularly useful in distinguishing species. For example, in some vespertilionids (Fenton and Bell, 1981; O’Farrell and Miller, 1999), and molossids (Fenton and Bell, 1981), lowest frequencies in calls are important, while in emballonurids, frequencies with most energy are particularly useful (Kalko, 1995). Otherauthors report that the frequency values measured at the ends of calls could be most useful in discriminating between species from different families (Parsons and Jones, 2000; Russo and Jones, 2002). However, the use of apparently diagnostic frequencies to identify bat species by their calls is complicated in species that vary frequencies between calls such as Saccopteryx bilineata and S. leptura...