Comparing distance-decay parameters: A novel test under pairwise dependence

Distance-decay models fit parametric functions to assess the relationship between similarity and spatial or environmental distance. Despite widespread use of distance-decay in ecology biogeography, no method has been previously developed validated significance differences parameters (i.e. intercept slope) two models. The pairwise autocorrelation distance affects variance parameter estimates, precluding ordinary t-tests. Here, we provide a test statistic (zdep) for equality can be applied, independently, any model and/or slopes). zdep accounts dependence, thus avoiding biases associated inflation degrees freedom, it is based on estimation parameters' using site-block resampling. To validate statistic, type I II errors were empirically evaluated through simulation study. We simulated six scenarios (three under null three alternative hypothesis) relationships different functions: negative exponential, power-law Gompertz function. applied computed proportion rejections hypothesis α = 0.01, 0.05 0.1 each scenario. As case-study, also compared across several groups Iberian vertebrates (cyprinids, frogs, lizards snakes, bats, rodents carnivores). In study, showed good approximation nominal level (α, error) statistical power (1 − error), later increasing with sample size, as expected. vertebrates, found significant ectotherms endotherms, but not within these except cyprinids other ectotherms. performance makes best option obtained from fitted data It used beyond approaches compare regression dependent (as genetic distances, example).