Featuring 10 phenological estimators using simulated data

摘要: Summary 1. Studies reporting phenological changes in response to climate change are numerous and concern all groups of living organisms. Phenological could cause mismatching food chains, thus inducing important perturbations ecosystem functioning. Nevertheless, the relevancy conclusions drawn from studies strongly depends on estimation accuracy such changes. Many different estimators exist some have already raised major criticism, although they continue be used. Therefore, there is a crucial need for an extensive study documenting behaviour estimators. 2. Here, we compare efficiency 10 estimators: first appearance dates, mean percentile dates smoothing method based spline functions using simulated data. Root mean-squared errors bias estimations calculated relation parameters data. 3. Results show that behave as very inaccurate biased estimator regarding any data set. Mean estimates provided general most accurate shifts. They were also robust variation sample sizes imperfect detectability. 4. Our results allow us warn against use future recommend or techniques estimate entire distributions. We provide advice concerning monitoring effort. These recommendations should importantly apply aiming at comparing among sites species.