Chasing flies: the identification of calyptrate flies using ecological surveys and wingbeat frequency as a species- and sex-specific classifier

Julie Pinto

Calyptrates is a species-rich clade that includes some of the most well-known taxa of flies such as blow flies (Calliphoridae), house flies (Muscidae), and flesh flies (Sarcophagidae). This clade also includes some of the insect world’s fastest and most agile flying insects. The accurate identification of these fly species is important for forensic, ecological, medical, and veterinary reasons. The focus of this thesis is the identification of these flies using two main approaches: (1) field surveys to morphologically identify and describe forensically relevant flies from a geographical area, and (2) the design and use of a novel technique to identify fly species using their wingbeat frequency. Two surveys on forensically relevant flies were completed in the north-eastern United States. In the first survey, five traps were set out seven times over a 2-year period, to collect adult and immature specimens at trap sites along an urban gradient. GIS mapping was used to classify a 1 km radius around each site using land use and the percent of urban impervious surfaces. Insect data collected from human corpses in this region were also tabulated to corroborate the forensic relevance of trapped specimens. A total of 4399 flies were collected, representing 5 genera and 19 species. Ninety-six percent of trapped adult specimens were represented by 3 species, all of which are widespread and synanthropic. The survey demonstrated that the forensically relevant flies of Central Connecticut have no habitat or seasonal preference, but rather temperature is the key determining factor affecting the number of flies trapped and the colonising species. The survey has provided much needed insect data for the geographical area. In the second survey, traps were placed weekly from June–September for 2 years in the city of New Haven, Connecticut, a city exposed to a prolonged period of urbanisation and industrialisation. Fly abundance and diversity data, climate, and land cover were compared to data collected 78 years earlier. The survey results suggest the adult fly population is now less diverse with decreased species richness and changes in the relative abundance of species. Species that showed a significant interaction with temperature in the 1940s survey have now increased in abundance. Several of the trapped species continue to show an interaction with temperature and rainfall. This study serves as a model for changes in the adult fly population in regions where highly urbanized cities have replaced natural environments. The use of wingbeat frequency to identify flying insects has been used in the surveillance and monitoring of insects within the agricultural and medical fields, but has yet to be exploited forensically and with calyptrate flies. The sound produced by the oscillation of insect wings during flight produces a sound wave consisting of a series of harmonics. The first harmonic is referred to as the wingbeat or fundamental frequency and is a feature that varies considerably across insect orders. Wingbeat frequency data of calyptrate flies is scant, with existing data only pertaining to flies that are either tethered or confined due to their fast and varied flight pattern. Recordings in such scenarios are further complicated by the low amplitude of the acoustic signal which attenuates rapidly with distance. These complexities were overcome by using a novel technique employing advances in sound editor and analysis software, making it possible to isolate low amplitude sounds using noise reduction and pitch detection algorithms. To demonstrate the identification of calyptrate flies using their wingbeat frequency, 40 (20 female, 20 male) specimens of 4 species: Lucilia sericata (Meigen), Calliphora dubia (Macquart), Musca vetustissima Walker, and Sarcophaga crassipalpis Macquart, were individually recorded in free flight at 26°C using a voice recorder. Fourier analysis was used to filter out noise, and an autocorrelation algorithm was used to estimate the mean fundamental wingbeat frequency of recorded specimens. The mean wingbeat frequency was found to be significantly different between species (F(3,156)=142.47, p<0.001), and discriminant analysis performed on the mean wingbeat frequency of the 3 carrion flies demonstrated that when sex is included, species were correctly classified 77.5% overall, with the correct identification of individual species ranging from 60% to 90%. Wingbeat frequency data collected in this study demonstrates the success of the methodology and forms the foundation of a future, larger dataset required for wingbeat frequency to identify forensically relevant calyptrate flies. The observed interspecific and intersexual differences may also indicate the use of wingbeat frequency by calyptrate male flies to recognise conspecific females. The relevance of these results is reviewed in the context of calyptrate communication and starts the discussion on the use of this metric by calyptrate flies as a species- and sex-specific communication cue.

Chasing flies: the identification of calyptrate flies using ecological surveys and wingbeat frequency as a species- and sex-specific classifier

Files and links (1)

Abstract

Details

Metrics