Skip to Main content Skip to Navigation
Journal articles

Convolutional neural networks improve species distribution modelling by capturing the spatial structure of the environment

Benjamin Deneu 1, 2 Maximilien Servajean 3 Pierre Bonnet 2, 4 Christophe Botella 1, 2 François Munoz 5 Alexis Joly 1
1 ZENITH - Scientific Data Management
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier, CRISAM - Inria Sophia Antipolis - Méditerranée
3 ADVANSE - ADVanced Analytics for data SciencE
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : Convolutional Neural Networks (CNNs) are statistical models suited for learning complex visual patterns. In the context of Species Distribution Models (SDM) and in line with predictions of landscape ecology and island biogeography, CNN could grasp how local landscape structure affects prediction of species occurrence in SDMs. The prediction can thus reflect the signatures of entangled ecological processes. Although previous machine-learning based SDMs can learn complex influences of environmental predictors, they cannot acknowledge the influence of environmental structure in local landscapes (hence denoted “punctual models”). In this study, we applied CNNs to a large dataset of plant occurrences in France (GBIF), on a large taxonomical scale, to predict ranked relative probability of species (by joint learning) to any geographical position. We examined the way local environmental landscapes improve prediction by performing alternative CNN models deprived of information on landscape heterogeneity and structure (“ablation experiments”). We found that the landscape structure around location crucially contributed to improve predictive performance of CNN-SDMs. CNN models can classify the predicted distributions of many species, as other joint modelling approaches, but they further prove efficient in identifying the influence of local environmental landscapes. CNN can then represent signatures of spatially structured environmental drivers. The prediction gain is noticeable for rare species, which open promising perspectives for biodiversity monitoring and conservation strategies. Therefore, the approach is of both theoretical and practical interest. We discuss the way to test hypotheses on the patterns learnt by CNN, which should be essential for further interpretation of the ecological processes at play.
Complete list of metadata

https://hal.inrae.fr/hal-03220977
Contributor : Yannick Brohard <>
Submitted on : Friday, May 7, 2021 - 4:07:08 PM
Last modification on : Thursday, May 27, 2021 - 10:49:43 AM

File

Deneu_etal_Plos_Computational_...
Publisher files allowed on an open archive

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Benjamin Deneu, Maximilien Servajean, Pierre Bonnet, Christophe Botella, François Munoz, et al.. Convolutional neural networks improve species distribution modelling by capturing the spatial structure of the environment. PLoS Computational Biology, Public Library of Science, 2021, 17 (4), pp.e1008856. ⟨10.1371/journal.pcbi.1008856⟩. ⟨hal-03220977⟩

Share

Metrics

Record views

84

Files downloads

62