Rigorous home-range estimation with movement data: a new autocorrelated kernel density estimator

Fleming and colleagues confronted a fundamental flaw in conventional kernel home-range estimation: it assumes location data are independent, yet modern GPS fixes are strongly autocorrelated in time, causing classic estimators to badly underestimate range area. They developed the autocorrelated kernel density estimator, which models the movement process explicitly and selects smoothing based on the data's correlation structure, yielding statistically defensible home-range estimates from dense, frequent fixes. The method, implemented in the ctmm framework, is now standard for analyzing high-resolution telemetry. For wide-ranging cats whose collars produce thousands of correlated fixes, AKDE corrects a systematic bias that had distorted earlier estimates. The paper marked a turning point toward home-range methods grounded in the underlying continuous-time movement process.