Environmental Conservation 37 (2): 147–154 C  Foundation for Environmental Conservation 2010 doi:10.1017/S0376892910000378 Patterning the distribution of threatened crayfish and their exotic analogues using self-organizing maps DOROTHÉE KOPP 1 , FRÉDÉRIC SANTOUL 1 , NICOLAS POULET 2 , ARTHUR COMPIN 1 AND RÉGIS CÉRÉGHINO 1 ∗ 1 EcoLab, UMR 5245, Université de Toulouse, bât 4R3, 118 route de Narbonne, 31062 Toulouse Cedex 9, France, and 2 Office National de l’Eau et des Milieux Aquatiques, 16 avenue Louison Bobet, 94132 Fontenay-sous-Bois, France Date submitted: 25 July 2009; Date accepted: 7 January 2010; First published online: 3 June 2010 SUMMARY Ability to demonstrate statistical patterns of distri- bution by threatened species and by their potential competitors will determine success in forecasting locations at greatest risk, and ability to target management efforts. A self-organizing map algorithm (SOM) was used to derive probabilities of presence of native (Austropotamobius pallipes) and exotic (Orconectes limosus, Pacifastacus leniusculus and Procambarus clarkii) crayfish species with respect to physical and land- cover variables in a large stream system, using a simple presence-absence dataset of species. Crayfish were sampled at 128 sites representing 86 rivers. The probability of occurrence of the native species increased at higher elevations above sea level and lower temperatures; populations appeared to be mostly confined to headwater streams where exotic competitors were unable to withstand the colder conditions. The distribution of exotic species was correlated with anthropogenic factors, such as the degree of urbanization and agricultural land area. Complementary modelling tools, such as GIS and SOMs, can help to maximize the information extracted from available data in the context of biological conservation. Keywords: freshwater, land use, neural networks, occurrence, species distribution INTRODUCTION At the onset of most action plans directed towards the conservation of threatened species at regional and/or national scales, numerical patterning is needed to ‘map’ the current distribution of populations in the area under survey, and, whenever possible, estimate abundances and densities (see Guisan & Zimmermann 2000 for a review). Eventually, similar information is required for non-native ecological analogues, when competitive interactions are likely to adversely affect the native species (Morgan 1998; de la Hoz Franco & Budy ∗ Correspondence: Professor Régis Céréghino Tel.: +33 561 558 436 e-mail: cereghin@cict.fr 2005). A common way to examine the fit between species distributions and habitat is through ordination and correlation of habitat and biotic variables. However, for many non- mutually exclusive reasons (for example time- and cost- efforts, species seasonality, detection difficulties and non- standardized sampling), quantitative data such as population densities cannot be consistently obtained over a large number of sampling sites (Margules & Austin 1994; Marshall et al. 2006), thus preventing conservationists from optimizing large but heterogeneous datasets built on the basis of field and/or literature data. There is therefore a need to develop alternative analytical approaches which can maximize the information extracted from available data, such as ‘simple’ presence- absence data (Manel et al. 1999; Bessa-Gomes & Petrucci- Fonseca 2003; Céréghino et al. 2005). Inspired by the structure and the mechanism of the human brain, artificial neural networks (ANNs) provide convenient tools to extract information from large ecological datasets (Lek & Guegan 2000). The self-organizing map (SOM; Kohonen 2001) is one of the most well-known unsupervised neural networks, performing a topology-preserving projection of the input data onto a regular two-dimensional space. In the output layer of the network, the neurons act as virtual samples and approximate the probability density function of the input data. Therefore, using a binary dataset of species occurrences, the SOM calculates quantitative continuous values which vary between 0 and 1, so that the occurrence probability of any species in a given area, in the form of the connection intensity, can be visualized onto a virtual map. Moreover, this technique is relevant to pattern detection in biological communities in relation to environmental data because the gradient distribution of some biological variables (for example species) can be visualized in a SOM previously trained with environmental variables only (Park et al. 2003), thus allowing the fit between a set of species and their environment to be examined. Amongst threatened aquatic animals, crayfish are the focus of many conservation studies in the northern hemisphere (Gil-Sánchez & Alba-Tercedor 2002; Renai et al. 2006; Trouilhé et al. 2007). Many factors closely related to fishing activities and/or human destruction of their physical and hydraulic habitats have led to declines (Light et al. 1995; Gil- Sánchez & Alba-Tercedor 2006). The white-clawed crayfish (Austropotamobius pallipes) is a listed species in Annex II of the European Community Habitats Directive 92/43/EEC,