Acoustic arrays can collect information from cryptic calling animals (e.g. frogs, crickets, gibbons, birds) without disturbing them. Novel statistical techniques developed at CREEM, St Andrews University allow estimation of the area sampled by the microphones, and estimates of the density of calling individuals. We are working with scientists at the Universities of Stellenbosch and Cape Town in South Africa, part-funded by the National Geographic Society to survey to deploy acoustic arrays at 400 sites across the Cape Peninsula Moss Frog range. The resulting spatial density models will allow us to investigate the two highest threats (fire and invasive trees) for which we have historical data going back 40 years.
Cape peninsula moss frog producing calls which are detected by two (A & C) of the four detectors represented by circles.
These frogs are so small that they are almost impossible to detect visually in their natural habitat. How can you find out how many there are? Their individual calls can be matched across multiple detectors (microphones) and analysed using spatial capture-recapture methods.
The mural shows one frog producing calls which are detected by two of the four detectors (microphones). The red circles represent those microphones that detected the calls while the yellow ones did not. Based on this information (detected on A and C, not detected on B and D) probability contour lines are produced about the location of the frog.
In an actual survey, several detectors would be placed throughout the study area. Using the location information about all detected frogs we can estimate how many frogs were in the study area.
Borchers, D.L., Stevenson, B.C., Kidney, D., Thomas, L.J. and Marques, T.A. 2014. A unifying model for capture-recapture and distance sampling surveys. Journal of the American Statistical Association 10: 195-204.
Stevenson, B. C., Borchers, D. L., Altwegg, R., Swift, R. J., Gillespie, D. M., and Measey, G. J. (2015) A general framework for animal density estimation from acoustic detections across a fixed microphone array. Methods in Ecology and Evolution, 6(1), 38-48.