BrianSMiller/callDensity: Bioacoustic call density estimation from single sensors in R v1.0.0

callDensity v1.0.0

First public release of callDensity, an R package for passive acoustic monitoring-based call density estimation using multi-observer capture-recapture methods.

This release accompanies the accepted manuscript:



Miller BS, Kelly N, Aulich MG, Balcazar N, Buchan SJ, Castro FR, Reeve K, Harris D. Common ground: efficient, consistent, observer-independent bioacoustic call density estimation with adjudicated ground truth and capture-recapture detection functions. Methods in Ecology and Evolution (in press).


Core functionality



Call density estimation via cde() using observer ground truth and adjudicated capture-recapture models

SNR-based detection function fitting for logistic (fitSNRdetectionFunc) and VGLM/VGAM (fitSNRvglm, fitSNRvgam) models

False discovery rate and CV estimation (falseDiscoveryRate, falseDiscoveryCV, ciFromCV)

Probability of detection in area via Monte Carlo simulation (pDetInArea), with optional parallel execution

Multi-observer capture history construction and manipulation


Supporting tools



Simulation framework for calls, acoustics, and detectors (simCallLocation, simCallAcoustics, simulateDetector)

Noise level estimation and SNR truncation

Diagnostic and results plots (densityPlot, detectionRatePlot, plotSNRHistogram, plotSpatialDetections)


Vignettes



Single and two-detector worked examples

Coastline integration for spatial detection plots

SNR threshold sensitivity analysis

Parallel execution benchmarks for pDetInArea