Classification of unlabeled observations in Species Distribution
Modelling using Point Process Models.
- Emy Guilbault,
- Ian Renner,
- Michael Mahony,
- Eric Beh
Ian Renner
The University of Newcastle Faculty of Science
Author ProfileMichael Mahony
The University of Newcastle Faculty of Science
Author ProfileAbstract
1. Species distribution modelling, which allows users to predict the
spatial distribution of species with the use of environmental
covariates, has become increasingly popular, with many software
platforms providing tools to fit species distribution models. However,
the species observations used in species distribution models can have
varying levels of quality and can have incomplete information, such as
uncertain species identity. 2. In this paper, we develop two algorithms
to reclassify observations with unknown species identities which
simultaneously predict different species distributions using spatial
point processes. We compare the performance of the different algorithms
using different initializations and parameters with models fitted using
only the observations with known species identity through simulations.
3. We show that performance varies with differences in correlation among
species distributions, species abundance, and the proportion of
observations with unknown species identities. Additionally, some of the
methods developed here outperformed the models that didn't use the
misspecified data. 4. These models represent an helpful and promising
tool for opportunistic surveys where misidentification happens or for
the distribution of species newly separated in their taxonomy.21 Oct 2020Submitted to Ecology and Evolution 23 Oct 2020Submission Checks Completed
23 Oct 2020Assigned to Editor
26 Oct 2020Review(s) Completed, Editorial Evaluation Pending
23 Dec 2020Editorial Decision: Revise Minor
08 Feb 20211st Revision Received
09 Feb 2021Submission Checks Completed
09 Feb 2021Assigned to Editor
09 Feb 2021Review(s) Completed, Editorial Evaluation Pending
11 Feb 2021Editorial Decision: Accept