An efficient extension of N-mixture models for multi-species abundance estimation

Juan P. Gomez, Scott K. Robinson, Jason K. Blackburn, José M. Ponciano, David Murrell
2017 Methods in Ecology and Evolution  
1 1. In this study we propose an extension of the N-mixture family of models 2 that targets an improvement of the statistical properties of rare species abun-3 dance estimators when sample sizes are low, yet typical size for tropical studies. 4 The proposed method harnesses information from other species in an ecological 5 community to correct each species' estimator. We provide guidance to deter-6 mine the sample size required to estimate accurately the abundance of rare 7 tropical species
more » ... attempting to estimate the abundance of single species. 8 1 peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/073577 doi: bioRxiv preprint first posted online Sep. 5, 2016; 2. We evaluate the proposed methods using an assumption of 50-m radius 9 plots and perform simulations comprising a broad range of sample sizes, true 10 abundances and detectability values and a complex data generating process. 11 The extension of the N-mixture model is achieved by assuming that the de-12 tection probabilities of a set of species are all drawn at random from a beta 13 distribution in a multi-species fashion. This hierarchical model avoids having 14 to specify a single detection probability parameter per species in the targeted 15 community. Parameter estimation is done via Maximum Likelihood. 16 3. We compared our multi-species approach with previously proposed multi-17 species N-mixture models, which we show are biased when the true densities 18 of species in the community are less than seven individuals per 100-ha. The 19 beta N-mixture model proposed here outperforms the traditional Multi-species 20 N-mixture model by allowing the estimation of organisms at lower densities 21 and controlling the bias in the estimation. 22 4. We illustrate how our methodology can be used to suggest sample sizes 23 required to estimate the abundance of organisms, when these are either rare, 24 common or abundant. When the interest is full communities, we show how 25 the multi-species approaches, and in particular our beta model and estimation 26 methodology, can be used as a practical solution to estimate organism densities 27 from rapid inventory datasets. The statistical inferences done with our model 28 via Maximum Likelihood can also be used to group species in a community 29 according to their detectabilities. 30
doi:10.1111/2041-210x.12856 pmid:29892335 fatcat:ouf4wzbmqbcvloh7u7a5x27wla