Spatial Downscaling of Alien Species Presences Using Machine Learning

Ioannis N Daliakopoulos,Stelios Katsanevakis,Aristides Moustakas

doi:10.3389/feart.2017.00060

Ioannis N Daliakopoulos, Stelios Katsanevakis + Show 1 more

Open Access

https://doi.org/10.3389/feart.2017.00060

Copy DOI

Abstract

Large scale, high-resolution data on alien species distributions are essential for spatially explicit assessments of their environmental and socio-economic impacts, and management interventions for mitigation. However, these data are often unavailable. This paper presents a method that relies on Random Forest (RF) models to distribute alien species presence counts at a finer resolution grid, thus achieving spatial downscaling. A sufficiently large number of RF models are trained using random subsets of the dataset as predictors, in a bootstrapping approach to account for the uncertainty introduced by the subset selection. The method is tested with an approximately 8×8 km2 grid containing floral alien species presence and several indices of climatic, habitat, land use covariates for the Mediterranean island of Crete, Greece. Alien species presence is aggregated at 16×16 km2 and used as a predictor of presence at the original resolution, thus simulating spatial downscaling. Potential explanatory variables included habitat types, land cover richness, endemic species richness, soil type, temperature, precipitation, and freshwater availability. Uncertainty assessment of the spatial downscaling of alien species’ occurrences was also performed and true/false presences and absences were quantified. The approach is promising for downscaling alien species datasets of larger spatial scale but coarse resolution, where the underlying environmental information is available at a finer resolution than the alien species data. Furthermore, the RF architecture allows for tuning towards operationally optimal sensitivity and specificity, thus providing a decision support tool for designing a resource efficient alien species census.

Highlights

The rate at which species are being translocated by humans beyond their native ranges, through a variety of pathways, has been accelerating (Essl et al, 2015)
Alien species pose a grave risk to biodiversity, ecosystem services, and human health, and their presence is an important constituent of the global change that we currently face (Vilà et al, 2011; Simberloff et al, 2013; Katsanevakis et al, 2014), there is an urgent need for targeted actions for prevention and mitigation
The least efficient node splits according to Mean Decrease Gini (MDG) were performed by artificial habitat richness, natural habitat richness, agricultural habitat richness, soil type richness, and temperature range (Figure 2, right)

Summary

Introduction

The rate at which species are being translocated by humans beyond their native ranges, through a variety of pathways, has been accelerating (Essl et al, 2015). The European Alien Species Information Network (EASIN; Katsanevakis et al, 2015), which has compiled the largest spatial dataset of alien species distribution in Europe, reports species presence data at a 10 × 10 km spatial resolution and for some species only at country level. Such coarse resolution is often inadequate for the needs of management and research, as data availability up to a point determines the outputs of the analysis in several ways including complexity, generality, utility, and predictive power (Evans et al, 2014; Evans and Moustakas, 2016). Accurate methods for downscaling coarse spatial data can be extremely useful in assessments of environmental and socio-economic impacts of alien species and in management interventions for mitigation

Methods

Results

Discussion

Conclusion