Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 3
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A comparison of artificial neural networks and random forests to predict native fish species richness in Mediterranean rivers

  • Olaya Marín, Esther Julia
  • Martinez-Capel, Francisco|||0000-0003-4991-0251
  • Vezza, Paolo
[EN] Machine learning (ML) techniques have become important to support decision making in management and conservation of freshwater aquatic ecosystems. Given the large number of ML techniques and to improve the understanding of ML utility in ecology, it is necessary to perform comparative studies of these techniques as a preparatory analysis for future model applications. The objectives of this study were (i) to compare the reliability and ecological relevance of two predictive models for fish richness, based on the techniques of artificial neural networks (ANN) and random forests (RF) and (ii) to evaluate the conformity in terms of selected important variables between the two modelling approaches. The effectiveness of the models were evaluated using three performance metrics: the determination coefficient (R2), the mean squared error (MSE) and the adjusted determination coefficient (R2adj) and both models were developed using a k-fold crossvalidation procedure. According to the results, both techniques had similar validation performance (R2 = 68% for RF and R2 = 66% for ANN). Although the two methods selected different subsets of input variables, both models demonstrated high ecological relevance for the conservation of native fish in the Mediterranean region. Moreover, this work shows how the use of different modelling methods can assist the critical analysis of predictions at a catchment scale., [FR] Les techniques d’apprentissage automatique (ML) sont devenues importantes pour aider à la décision dans la gestion et la conservation des écosystèmes aquatiques d’eau douce. Étant donné le grand nombre de techniques ML pour améliorer la compréhension de l’utilité des ML en écologie, il est nécessaire de réaliser des études comparatives de ces techniques comme analyse préparatoire pour des applications de modèles futurs. Les objectifs de cette étude étaient : (i) de comparer la fiabilité et la pertinence écologique de deux modèles prédictifs pour la richesse de poisson, basé sur les techniques de réseaux de neurones artifi- ciels (ANN) et les forêts aléatoires (RF) et (ii) d’évaluer la conformité en termes de sélection des variables importantes entre les deux approches de modélisation. L’efficacité des modèles a été évaluée au moyen de trois indicateurs de performance : le coefficient de détermination (R2), l’erreur quadratique moyenne (MSE) et le coefficient de détermination ajusté (R2 adj) et les deux modèles ont été développés en utilisant une procédure de validation croisée k-fold. Selon les résultats, les deux techniques ont des performances de validation similaires (R2 = 68 % pour RF et R2 = 66 % pour ANN). Bien que les deux méthodes aient choisi différents sous-ensembles de variables d’entrée, les deux modèles ont démontré la pertinence écologique pour la conservation des poissons indigènes dans la région méditerranéenne. En outre, ce travail montre comment l’utilisation de différentes méthodes de modélisation peut aider à l’analyse critique des prévisions à l’échelle du bassin versant., This study was partially funded by the Spanish Ministry of Economy and Competitiveness with the projects SCARCE (Consolider-Ingenio 2010 CSD2009-00065) and POTECOL "Evaluacion del Potencial Ecologico de R os Regulados por Embalses y Desarrollo de Criterios para su mejora segun la Directiva Marco del Agua" (CGL2007-66412). In addition, the RF analysis was developed in the frame of the EU-funded HolRiverMed project (IEF, Marie Curie Actions). We thank the Confederacion Hidrografica del Jucar (Spanish Ministry of Agriculture, Food and Environment) for the data provided to develop this study and we also owe our gratitude to Sasa Plestenjak for the collaboration in building the first fish database for this research. We owe our gratitude to Chris Holmquist-Johnson and Leanne Hanson (USGS, Fort Collins Science Center) for the scientific review of the paper.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Modelling critical factors affecting the distribution of the vulnerable endemic Eastern Iberian barbel (Luciobarbus guiraonis) in Mediterranean rivers

  • Olaya Marín, Esther Julia
  • Martinez-Capel, Francisco|||0000-0003-4991-0251
  • García-Bartual, Rafael|||0000-0002-1811-8264
  • Vezza, Paolo
[EN] Luciobarbus guiraonis (Eastern Iberian barbel) is an endemic fish species restricted to Spain, mainly distributed in the Júcar River Basin District. Its study is important because there is little knowledge about its biology and ecology. To improve the knowledge about the species distribution and habitat requirements, nonlinear modelling was carried out to predict the presence/absence and density of the Eastern Iberian barbel, based on 155 sampling sites distributed throughout the Júcar River Basin District (Eastern Iberian Peninsula). We used multilayer feed-forward artificial neural networks (ANN) to represent nonlinear relationships between L. guiraonis descriptors and variables regarding the physical habitat and biological components (macroinvertebrates, fish, riparian forest). The gradient descent algorithm was implemented to find the optimal model parameters; the importance of the ANN s input variables was determined by the partial derivatives method. The predictive power of the model was evaluated with the Cohen s kappa (k), the correctly classified instances (CCI), and the area under the curve (AUC) of the receiver operator characteristic (ROC) plots. The best model predicted presence/absence with a high performance (k= 0.66, CCI= 87% and AUC= 0.85); the prediction of density was moderate (CCI = 62%, AUC=0.71 and k= 0.43). The fundamental variables describing the presence/absence were; solar radiation (the highest contribution was observed between 2000 and 4200 WH/m2), drainage area (with the strongest influence between 3000 and 5.000 km2), and the proportion of exotic fish species (with relevant contribution between 50 and 100%). In the density model, the most important variables were the coefficient of variation of mean annual flows (relative importance of 50.5%) and the proportion of exotic fish species (24.4%). The models provide important information about the relation of L. guiraonis with biotic and abiotic variables, this new knowledge can help develop future studies and management plans for the conservation of this species in the Júcar River Basin District and, potentially, for the conservation of other endemic fish species of Barbus and Luciobarbus in Mediterranean rivers., This study was partially funded by the MINECO, Spanish Ministry of Economy and Competitiveness, with the projects SCARCE (Consolider-Ingenio 2010 CSD2009-00065), IMPADAPT (CGL2013-48424- C2-1-R) and with Feder funds. There was also funding from the Project POTECOL “Evaluación del Potencial Ecológico de Ríos Regulados por Embalses y Desarrollo de Criterios para su mejora según la Directiva Marco del Agua” (CGL2007-66412). In addition, the present research was developed in the frame of the EU-funded HolRiverMed project (Marie Curie Actions, IntraEuropean Fellowships for the mobility of European researchers). We thank the Confederación Hidrográfica del Júcar (Spanish Ministry of Agriculture, Food and Environment) for the data provided to develop this study and we also owe our gratitude to Juan Jiménez for the collaboration in building the first fish database for this research.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Random forests to evaluate interspecific interactions in fish distribution models

  • Vezza, Paolo
  • Martinez-Capel, Francisco|||0000-0003-4991-0251
  • Muñoz Mas, Rafael
  • Mouton, A.
[EN] Previous research indicated that high predictive performance in species distribution modelling can be obtained by combining both biotic and abiotic habitat variables. However, models developed for fish often only address physical habitat characteristics, thus omitting potentially important biotic factors. Therefore, we assessed the impact of biotic variables on fish habitat preferences in four selected stretches of the upper Cabriel River (E Spain). The occurrence of Squalius pyrenaicus and Luciobarbus guiraonis was related to environmental variables describing biotic interactions (inferred by relationships among fish abundances) and channel hydro-morphological characteristics. Random Forests (RF) models were trained and then validated using independent datasets. To build RF models, the conditional variable importance was used together with the model improvement ratio technique. The procedure showed effectiveness in identifying a parsimonious set of not correlated variables, which minimize noise and improve model performance in both training and validation phases. Water depth, channel width, fine substrate and water-surface gradient were selected as most important habitat variables for both fish. Results showed clear habitat overlapping between fish species and suggest that competition is not a strong factor in the study area., This research has been developed in the framework of the HolRiverMed project (FP7-PEOPLE-2010-275577, Marie Curie Actions, Intra-European Fellowships) and the SCARCE project (Assessing and predicting effects on water quantity and quality in Iberian rivers caused by global change, Consolider-Ingenio 2010 CSD2009-00065). Data collection was partially funded by the Spanish Ministry of Environment, Rural and Marine Affairs, the Jucar River Basin District Authority and the Spanish Ministry of Education and Science (POTECOL, CGL2007-66412). We thank Juan Diego Alacaraz-Hernandez, Matias Peredo-Parada and Aina Hernandez-Mascarell for their help with field work and suggestions on data analysis.