Generalized additive models to predict adult and young brown trout (<i>Salmo trutta</i>Linnaeus, 1758<i>)</i>densities in Mediterranean rivers

Alcaraz‐Hernández, Juan Diego; Muñoz‐Mas, Rafael; Martínez‐Capel, Francisco; Garófano‐Gómez, Virginia; Vezza, Paolo

doi:10.1111/jai.13025

Cited by 9 publications

(14 citation statements)

References 88 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Three major interrelated features requiring optimization control GAMs' performance: the number of selected inputs, overparameterization, which refers to the excessive number of effective degrees of freedom (Alcaraz‐Hernández, Muñoz‐Mas, Martínez‐Capel, Garófano‐Gómez, & Vezza, ), and data prevalence, which corresponds to the ratio of presence data within the entire dataset (Platts, McClean, Lovett, & Marchant, ). Therefore, a wrapper approach (see, e.g., Muñoz‐Mas, Fukuda, Vezza, & Martínez‐Capel, ), based on cross‐validation, genetic algorithm (GA), and evolutionary algorithm (Holland, ), was used to obtain the most relevant noncorrelated ( r 2 > 0·66; p ≤ 0·1) input variables (Peters et al, ), the optimal number of effective degrees of freedom (Arlot & Celisse, ), and the case weights, in order to balance the error committed on presence and absence data (Muñoz‐Mas, Lopez‐Nicolas, et al, ).…”

Section: Methodsmentioning

confidence: 99%

Exploring the key drivers of riparian woodland successional pathways across three European river reaches

et al. 2017

Self Cite

View full text Add to dashboard Cite

International audienceClimate change and river regulation are negatively impacting riparian vegetation. To evaluate these impacts, process‐based models are preferred over data‐driven approaches. However, they require extensive knowledge about ecohydrological processes. To facilitate the implementation of such process‐based models, the key drivers of riparian woodland successional pathways across three river reaches, in Austria, Portugal, and Spain, were explored, employing two complementary approaches. The principal component analyses highlighted the importance of the physical gradients determining the placement of the succession phases within the riparian and floodplain zones. The generalized additive models revealed that the initial and pioneer succession phases, characteristic of the colonization stage, appeared in areas highly morphodynamic, close in height and distance to the water table, and with coarse substrate, whereas elder phases within the transitional and mature stages showed incremental differences, occupying less dynamic areas with finer substrate. The Austrian site fitted well the current successional theory (elder phases appearing sequentially further up and distant), but at the Portuguese site, the tolerance of the riparian species to drought and flash flood events governed their placement. Finally, at the Spanish site, the patchy distribution of the elder phases was the remnants of formative events that reshaped the river channel. These results highlight the complex relationships between flow regime, channel morphology, and riparian vegetation. The use of succession phases, which rely on the sequential evolution of riparian vegetation as a response to different drivers, may be potentially better reproducible, within numerical process‐based models, and transferable to other geographical regions

show abstract

Section: Methodsmentioning

confidence: 99%

Exploring the key drivers of riparian woodland successional pathways across three European river reaches

et al. 2017

Self Cite

View full text Add to dashboard Cite

show abstract

“…Because microhabitat selection is typically a nonlinear process (e.g., Alcaraz-Hernandez, Munoz-Mas, Martinez-Capel, Garofano-Gomez, & Vezza, 2016;Girard et al, 2014;Labonne et al, 2003), we introduced B-splines to transform the microhabitat variables, which decomposed these variables into piecewise cubic regressions with fixed knots (Grajeda et al, 2016;Pan & Goldstein, 1998). For each model, we previously selected the appropriate number of knots between models with a single knot (fixed at 50% of the distribution of the microhabitat variable, two degrees of freedom) or models with two knots (fixed at 33 and 67% of the distribution of the microhabitat variable, three degrees of freedom).…”

Section: Modellingmentioning

confidence: 99%

Predictive models of fish microhabitat selection in multiple sites accounting for abundance overdispersion

Plichard

Forcellini

Coarer

et al. 2020

River Research & Apps

View full text Add to dashboard Cite

Microhabitat selection models are frequently used in rivers to evaluate anthropogenic effects on aquatic organisms. Fish models are generally developed from few rivers, with debatable statistical treatments for coping with overdispersed abundance distributions. Analyses of data from multiple rivers are needed to test their transferability and increase their relevance for stakeholders. Using 3,528 microhabitats sampled in nine French rivers during 129 surveys, we developed models for 35 specific size classes of 22 fish species. We used mixed‐effects generalized linear models (accounting for multiple surveys), involving B‐spline transformations (accounting for nonlinear responses) and assuming a negative binomial distribution (accounting for abundance overdispersion). We compared models of increasing complexity: no selection (M1), an “average” selection similar in all surveys (M2), two models with different selection across surveys (M3–M4). Of 132 univariate cases (specific size classes by habitat), 63% indicated selection for depth, 71% for velocity, 45% for substratum size and 13% for substratum heterogeneity. A total of 50 models were retained, involving 26/35 specific size classes. Model fits indicated low explained deviance (R2MF < 0.19) and higher rank correlations (ρ < 0.69) between observed and modelled values. However, Bayesian posterior predictive checks validated these results since excellent fits would generate R2MF lower than 0.59 and ρ lower than 0.78. We found high transferability among rivers and dates, because (a) M2 was the most appropriate in 26/50 cases; (b) the R2MF and ρ values by M2 was, respectively, 72% and 75% of that explained by the complex M4 and (c) independent river cross‐validations showed good transferability. Bivariate models for selected specific size classes improved univariate model fits (ρ from 0.30 to 0.38). Overall, using a nonlinear mixed‐effect approach, our results confirmed the relevance of “average” models based on several rivers for developing helpful e‐flow tools. Finally, our modelling approach opens opportunities for integrating additional effects as the spatial distribution of competitors.

show abstract

“…() showed that elevation combined with local habitat features (e.g., cover) accounted for much of the variation of brown trout biomass and density across 264 stream reaches belonging to 15 streams. Local habitat attributes such as the mean velocity, the cover, the maximum depth, the distance between rapids, and the number of slow habitats were used together with the reach elevation to model young (<2 years) brown trout densities in the Júcar River Basin District (Spain; Alcaraz‐Hernández et al ., ). Depending on the local morphological and hydraulic characteristics, size‐class habitat selection patterns by brown trout exhibit large flexibility and adapt to local conditions (Ayllón, Almodóvar, Nicola, & Elvira, ), suggesting that the influence of a given habitat variable is site‐dependent.…”

Section: Introductionmentioning

confidence: 99%

“…Variation in space in trout population abundance, structure, or production are common, both within and among rivers (Milner, Gee, & Hemsworth, 1979;Milner, Wyatt, & Scott, 1993;Cattanéo et al, 2002;Lobón-Cerviá & Rincón, 2004;Almodóvar, Nicola, & Elvira, 2006;Lobón-Cerviá, 2007;Zorn & Nuhfer, 2007;Richard, Cattanéo, & Rubin, 2015). These dynamics can be related to catchment scale attributes (e.g., basin area, stream order, altitude, and geology) and climate, but much of this variability is explained by local scale attributes (i.e., site; e.g., depth, substrate site, cover, and water velocity, Milner, Hemsworth, & Jones, 1985;Armstrong, Kemp, Kennedy, Ladle, & Milner, 2003;Santiago et al, 2016;Alcaraz-Hernández, Muñoz-Mas, Martínez-Capel, Garófano-Gómez, & Vezza, 2016). In the French Pyrénées, Baran, Delacoste, Lascaux, and Belaud (1993); Baran, Delacoste, Poizat et al (1995) showed that elevation combined with local habitat features (e.g., cover) accounted for much of the variation of brown trout biomass and density across 264 stream reaches belonging to 15 streams.…”

mentioning

confidence: 99%

Hydrological drivers of brown trout population dynamics in France

Bergerot

Cattanéo

2016

Ecohydrology

View full text Add to dashboard Cite

International audienceAlteration of natural river flow regimes is a major threat to freshwater biodiversity. Restoration of natural flow regimes requires an understanding of linked hydrological and ecological processes. In this study, we investigated how annual seasonal flow characteristics and physical habitat attributes could interact to explain yearly changes in brown trout population densities (for 0+ fish, corresponding to the young of the year, and for the >0+ age class) in 112 sites widespread in France. Using an information theoretic approach and general linear modeling, we identified the physical habitat attributes and seasonal hydrological variables which explained 0+ proportion, total, >0+ and 0+ brown trout density dynamics. A decrease in total and 0+ brown trout densities were mainly linked to high water levels in both high and low flows during the emergence period and to a lesser extent, to physical habitat attributes reflecting the river size. 0+ proportions were only linked to the level of low flows, and to a lesser extent to high flows, during the emergence period. Our study demonstrated that seasonality of flow is a main driver of brown trout population dynamics, especially magnitude of high and low flows for the 0+ age class, and should be considered in the definition of environmental flows

show abstract

Generalized additive models to predict adult and young brown trout (Salmo truttaLinnaeus, 1758)densities in Mediterranean rivers

Abstract: Wiley: 12 months Alcaraz-Hernández, JD.; Muñoz Mas, R.; Martinez-Capel, F.; Garófano-Gómez, V.; Vezza, P. (2016). Generalized additive models to predict adult and young brown trout (Salmo trutta Linnaeus, 1758) 2

Cited by 9 publications

References 88 publications

Exploring the key drivers of riparian woodland successional pathways across three European river reaches

Exploring the key drivers of riparian woodland successional pathways across three European river reaches

Predictive models of fish microhabitat selection in multiple sites accounting for abundance overdispersion

Hydrological drivers of brown trout population dynamics in France

Contact Info

Product

Resources

About