Not seeing the forest for the trees: Generalised linear model out-performs random forest in species distribution modelling for Southeast Asian felids

Chiaverini，Luca1; Macdonald，David W.1; Hearn，Andrew J.1; Kaszta，Żaneta1; Ash，Eric1,2; Bothwell，Helen M.1,3; Can，Özgün Emre4; Channa，Phan1,5; Clements，Gopalasamy Reuben6,7,8; Haidir，Iding Achmad1,9; Kyaw，Pyae Phyoe1,10; Moore，Jonathan H.11,12; Rasphone，Akchousanh1,13; Tan，Cedric Kai Wei1,14; Cushman，Samuel A.1,15

2023-07-01

10.1016/j.ecoinf.2023.102026

Ecological Informatics   影响因子和分区

1574-9541

1878-0512

Species Distribution Models (SDMs) are a powerful tool to derive habitat suitability predictions relating species occurrence data with habitat features. Two of the most frequently applied algorithms to model species-habitat relationships are Generalised Linear Models (GLM) and Random Forest (RF). The former is a parametric regression model providing functional models with direct interpretability. The latter is a machine learning non-parametric algorithm, more tolerant than other approaches in its assumptions, which has often been shown to outperform parametric algorithms. Other approaches have been developed to produce robust SDMs, like training data bootstrapping and spatial scale optimisation. Using felid presence-absence data from three study regions in Southeast Asia (mainland, Borneo and Sumatra), we tested the performances of SDMs by implementing four modelling frameworks: GLM and RF with bootstrapped and non-bootstrapped training data. With Mantel and ANOVA tests we explored how the four combinations of algorithms and bootstrapping influenced SDMs and their predictive performances. Additionally, we tested how scale-optimisation responded to species' size, taxonomic associations (species and genus), study area and algorithm. We found that choice of algorithm had strong effect in determining the differences between SDMs' spatial predictions, while bootstrapping had no effect. Additionally, algorithm followed by study area and species, were the main factors driving differences in the spatial scales identified. SDMs trained with GLM showed higher predictive performance, however, ANOVA tests revealed that algorithm had significant effect only in explaining the variance observed in sensitivity and specificity and, when interacting with bootstrapping, in Percent Correctly Classified (PCC). Bootstrapping significantly explained the variance in specificity, PCC and True Skills Statistics (TSS). Our results suggest that there are systematic differences in the scales identified and in the predictions produced by GLM vs. RF, but that neither approach was consistently better than the other. The divergent predictions and inconsistent predictive abilities suggest that analysts should not assume machine learning is inherently superior and should test multiple methods. Our results have strong implications for SDM development, revealing the inconsistencies introduced by the choice of algorithm on scale optimisation, with GLM selecting broader scales than RF.

Bootstrapping Generalised linear model Machine learning Multi-scale Random forest Species distribution modelling

SCI

英语

其他

Environmental Sciences & Ecology

Ecology

WOS:000948673200001

ELSEVIER

2-s2.0-85149993211

Scopus

1.Wildlife Conservation Research Unit,Department of Biology,University of Oxford,The Recanati-Kaplan Centre,Tubney House,Tubney,Oxon,OX13 5QL,United Kingdom
2.Freeland Foundation,Bangkok,Thailand
3.Research School of Biology,Australian National University,Canberra,Australia
4.Department of Biology,Faculty of Science,Ankara University,Ankara,Turkey
5.General Directorate of Natural Protected Area,Ministry of Environment,Phnom Penh,Cambodia
6.Rimba,Kuala Lumpur,Malaysia
7.Department of Biological Sciences,Sunway University,Bandar Sunway,Malaysia
8.Jeffrey Sachs on Sustainable Development,Sunway University,Bandar Sunway,Malaysia
9.Directorate of Conservation Area Planning,Directorate General of Natural Resources and Ecosystem Conservation,Ministry of Environment and Forestry,Jakarta,Indonesia
10.Wildlife Conservation Society – Myanmar Program,Yangon,Myanmar
11.School of Environmental Science and Engineering,Southern University of Science and Technology,Shenzhen,China
12.School of Environmental Sciences,University of East Anglia,Norwich,United Kingdom
13.Wildlife Conservation Society – Lao PDR Program,Vientiane,Laos
14.School of Environmental and Geographical Sciences,University of Nottingham Malaysia,Semenyih,Malaysia
15.Rocky Mountain Research Station,United States Forest Service,Flagstaff,United States

Chiaverini，Luca,Macdonald，David W.,Hearn，Andrew J.,et al. Not seeing the forest for the trees: Generalised linear model out-performs random forest in species distribution modelling for Southeast Asian felids[J]. Ecological Informatics,2023,75.

Chiaverini，Luca.,Macdonald，David W..,Hearn，Andrew J..,Kaszta，Żaneta.,Ash，Eric.,...&Cushman，Samuel A..(2023).Not seeing the forest for the trees: Generalised linear model out-performs random forest in species distribution modelling for Southeast Asian felids.Ecological Informatics,75.

Chiaverini，Luca,et al."Not seeing the forest for the trees: Generalised linear model out-performs random forest in species distribution modelling for Southeast Asian felids".Ecological Informatics 75(2023).