Towards combining occurrence and abundance distribution models of Great Bustard for conservation: A global research template from Bohai Bay?
- Published
- Accepted
- Subject Areas
- Biodiversity, Biogeography, Conservation Biology, Ecology, Data Mining and Machine Learning
- Keywords
- conservation decision, Great Bustard (Otis tarda dybowskii), abundance model, Random Forest, occurrence model, machine learning method
- Copyright
- © 2017 Mi et al.
- Licence
- This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Preprints) and either DOI or URL of the article must be cited.
- Cite this article
- 2017. Towards combining occurrence and abundance distribution models of Great Bustard for conservation: A global research template from Bohai Bay? PeerJ Preprints 5:e3240v1 https://doi.org/10.7287/peerj.preprints.3240v1
Abstract
Species distribution models (SDMs) have become important and essential tools in conservation and management. However, SDMs built with count data, commonly referred to as species abundance models (SAMs), are still less used so far. SDMs are increasingly used now in conservation decisions, whereas SAMs are still not widely employed. Species occurrence and abundance do not frequently display similar patterns, often they are not even well correlated. This leads to an insufficient or misleading conservation. How to combine information from SDMs and SAMs all together for unified conservation remains a challenge. In this study, we put forward for the first time a priority protection index (PI). The PI combines the prediction results of occurrence and abundance models. We used the best-available presence and count records for an endangered farmland species, Great Bustard (Otis tarda dybowskii) in Bohai Bay, China, as a case study. We then applied the advanced Random Forest algorithm (Salford Systems Ltd. implementation), a powerful machine learning method, with eleven predictor variables to forecast the spatial occurrence as well as the abundance distribution. The results show that the occurrence model had a decent performance (ROC: 0.77) and the abundance model had a RMSE 26.54. It is of note that environmental variables influenced bustard occurrence and abundance differently. We found that occurrence and abundance models display different spatial distribution patterns. Still, combining occurrence and abundance indices to produce a priority protection index (PI) used for conservation could guide the protection of the areas with high occurrence and high abundance (e.g. in Strategic Conservation Planning). Due to the widespread use of SDMs and the rel. easy subsequent employment of SAMs these findings have a wide relevance and applicability, worldwide. We promote and strongly encourage to further test, apply and update the priority protection index (PI) elsewhere in order to explore the generality of these findings and methods readily available now for researchers.
Author Comment
This is a submission to PeerJ for review.