Review and comparison of methods to study the contribution of variables in artificial neural network models

Convinced by the predictive quality of artificial neural network (ANN) models in ecology, we have turned our interests to their explanatory capacities. Seven methods which can give the relative contribution and/or the contribution profile of the input factors were compared: (i) the ‘PaD’ (for Partial Derivatives) method consists in a calculation of the partial derivatives of the output according to the input variables; (ii) the ‘Weights’ method is a computation using the connection weights; (iii) the ‘Perturb’ method corresponds to a perturbation of the input variables; (iv) the ‘Profile’ method is a successive variation of one input variable while the others are kept constant at a fixed value; (v) the ‘classical stepwise’ method is an observation of the change in the error value when an adding (forward) or an elimination (backward) step of the input variables is operated; (vi) ‘Improved stepwise a’ uses the same principle as the classical stepwise, but the elimination of the input occurs when the network is trained, the connection weights corresponding to the input variable studied is also eliminated; (vii) ‘Improved stepwise b’ involves the network being trained and fixed step by step, one input variable at its mean value to note the consequences on the error. The data tested in this study concerns the prediction of the density of brown trout spawning redds using habitat characteristics. The PaD method was found to be the most useful as it gave the most complete results, followed by the Profile method that gave the contribution profile of the input variables. The Perturb method allowed a good classification of the input parameters as well as the Weights method that has been simplified but these two methods lack stability. Next came the two improved stepwise methods (a and b) that both gave exactly the same result but the contributions were not sufficiently expressed. Finally, the classical stepwise methods gave the poorest results.     

Habitat Loss and Changes in the Species-Area Relationship

Abstract: The species-area relationship (SAR) has been used successfully to predict extinction from extent of habitat reduction. These extinction estimates assume that species have uniformly distributed range requirements and a minimum abundance level required for persistence; how many species are lost depends solely on how much habitat is removed, not on where it is removed. We consider another limiting case in which range requirements, rather than abundances, determine extinctions. We used a new method for constructing SARs based on assumptions about geographic ranges of species. Our results show that habitat destruction can change the SAR and consequently the number of species predicted to be lost due to habitat destruction. Our method generates SARs that vary in shape according to the specific distributions of geographic range and occupancy but that have the common feature of being described by a power law with an exponent of <1. When the geographic range of species was included in the SAR, the way habitat was lost became important. Although the SAR before habitat destruction is often used to predict species loss after habitat destruction, assumptions must be clearly stated. To predict the damage caused by habitat loss with our model, it is necessary to know the fraction of aggregated species, the distribution of geographic ranges, the form of habitat destruction, and the sampling protocol. The remaining theoretical challenge is to develop a full theory that links abundance and range.     

Response of phase II detoxification enzymes in Phragmites australis plants exposed to organochlorines

Mixed pollution is a characteristic of many industrial sites and constructed wetlands. Plants possessing an enzymatic detoxifying system that is able to handle xenobiotics seems to be a viable option for the removal of mixed persistent contaminants such organochlorines (OCs: monochlorobenzene (MCB), 1,4-dichlorobenzene (DCB), 1,2,4-trichlorobenzene (TCB), γ-hexachlorocyclohexane (HCH)). In this study, Phragmites australis plants were exposed to sub-lethal concentrations of OCs (7 days), in single-exposure (0.8 to 10 mg?l^?1) and in mixture of OCs (0.2 mg?l^?1 MCB?+?0.2 mg?l^?1 DCB?+?2.5 mg?l^?1 TCB?+?0.175 mg?l^?1 HCH). Studies were conducted on the detoxification phase II enzymes; glutathione S-transferases (GST), and glucosyltransferases (UGT). Measurements of GST and UGT activities revealed that OCs may be buffered by glutathione and glucose conjugation. There appeared to be a correlation between the effects on phase II enzymes and the degree of chlorination of the benzene ring with, for example, the greatest effects being obtained for HCH exposure. In the case of mixed pollution, the induction of some GST isoenzymes (CDNB, 35 % non-significant) and UGT (118 %) in leaves and the inhibition of phase II enzymes in the other organs were measured. UGTs appear to be key enzymes in the detoxification of OCs.     

Application Of Artificial Neural Network Models To Analyse The Relationships Between Gammarus pulex L. (Crustacea, Amphipoda) And River Characteristics

This study aimed at analysing the relationship between river characteristics and abundance of Gammarus pulex. To this end, four methods which can identify the relative contribution and/or the contribution profile of the input variables in neural networks describing the habitat preferences of this species were compared: (i) the ‘PaD‘ (‘Partial Derivatives‘) method consists of a calculation of the partial derivatives of the output in relation to the input variables; (ii) the ‘Weights‘method is a computation using the connection weights of the backpropagation Artificial Neural Networks; (iii) the ‘Perturb‘method analyses the effect of a perturbation of the input variables on the output variable; (iv) the ‘Profile‘ method is a successive variation of one input variable while the others are kept constant at a fixed set of values. The dataset consisted of 179 samples, collected over a three-year period in the Zwalm river basin in Flanders, Belgium. Twenty-four environmental variables as well as the log-transformed abundance of Gammarus pulex were used in this study. The different contribution methods gave similar results concerning the order of importance of the input variables. Moreover, the stability of the methods was confirmed by gradually removing variables. Only in a limited number of cases a shift in the relative importance of the remaining input variables could be observed. Nevertheless, differences in sensitivity and stability of the methods were detected, probably as a result of the different calculation procedures. In this respect, the ‘PaD‘method made a more severe discrimination between minor and major contributing environmental variables in comparison to the ‘Weights‘, ‘Profile‘ and ‘Perturb‘ methods. From an ecological point of view, the input variables ‘Ammonium‘ and to a smaller extent ‘COD‘, were selected by these methods as dominant river characteristics for the prediction of the abundance of Gammarus pulex in this study area.     

Nitrogen Removal by Stormwater Management Structures: A Data Synthesis

A comprehensive synthesis of data from empirically based published studies and a widely used stormwater best management practice (BMP) database were used to assess the variability in nitrogen (N) removal performance of urban stormwater ponds, wetlands, and swales and to identify factors that may explain this variability. While the data suggest that BMPs were generally effective on average, removal efficiencies of ammonium (NH₄), nitrate (NO₃), and total nitrogen (TN) were highly variable ranging from negative (i.e., BMPs acting as sources of N) to 100%. For example, removal of NO₃ varied from (median ±1 SD) ?15 ± 49% for dry ponds, 32 ± 120% for wet ponds, 58 ± 210% for wetlands, and 37 ± 29% for swales. Across the same BMP types, TN removal was 27 ± 24%, 40 ± 31%, 61 ± 30%, and 50 ± 29%. NH₄ removal was 9 ± 36%, 29 ± 72%, 31 ± 24%, and 45 ± 34%. BMP size, age, and location explained some of the variability. For example, small and shallow ponds and wetlands were more effective than larger, deeper ones in removing N. Despite well‐known intra‐annual variation in N fluxes, most measurements have been made over short time periods using concentrations, not flow‐weighted N fluxes. Urban N export is increasing in some areas as large storms become more frequent. Thus, accounting for the full range of BMP performance under such conditions is crucial. A select number of long‐term flux‐based BMP studies that rigorously measure rainfall, hydrology, and site conditions could improve BMP implementation.     

Turbulence characteristics within sparse and dense canopies

Boundary layer interactions with canopies control various environmental processes. In the case of dense and homogeneous canopies, the so-called mixing layer analogy is most generally used. When the canopy becomes sparser, a transition occurs between the mixing layer and the boundary layer perturbed by interactions between element wakes. This transition has still to be fully understood and characterized. The experimental work presented here deals with the effect of the canopy density on the flow turbulence and involves an artificial canopy placed in a fully developed turbulent boundary layer. One and two-component velocity measurements are performed, both within and above the canopy. The influence of the spacing between canopy elements is studied. Longitudinal velocity statistical moments and Reynolds stresses are calculated and compared to literature data. For spacings greater than the canopy height, evidences of this transition are found in the evolution of the skewness factor, shear length scale and mixing length.     

Two unusual chromosome aberrations ascertained by sonographic anomalies

We describe two cases of sonographic abnormalities associated with unusual chromosomal aberrations. Case 1 presented with a cystic hygroma at 12 weeks' gestation. Cytogenetic analysis revealed an unbalanced complex chromosome rearrangement implicating chromosomes 6, 13 and 21 (karyotype: 47,XX,t(6;21;14)(q14;q21;q21)mat,+21) and corresponding to a complete trisomy 21. This anomaly resulted from malsegregation of a maternal balanced three-way translocation. For case 2, an alobar holoprosencephaly was identified by ultrasonography at 23 weeks' gestation. Chromosomal analysis showed a recombinant rec (13), dup q chromosome, secondary to unequal crossing-over of a paternal pericentric inversion of chromosome 13, giving rise to partial trisomy 13q (karyotype: 46,XX,rec(13)dup(13q)inv(13)(p11q21)pat). These two cases illustrate the role of ultrasound in leading to detection not only of foetal chromosomal aberrations but also of rare balanced chromosomal rearrangements presented by one of the two parents. Copyright © 2004 John Wiley & Sons, Ltd.     

Removal of estrogenic activity of natural and synthetic hormones from a municipal wastewater: efficiency of horseradish peroxidase and laccase from Trametes versicolor

Some researches studied the removal of steroid estrogens by enzymatic treatment, however none verified the residual estrogenicity after the enzymatic treatment at environmental conditions. In this study, the residual estrogenic activities of the key natural and synthetic steroid estrogens were investigated following enzymatic treatment with horseradish peroxidase (HRP) and laccase from Trametes versicolor. Synthetic water and municipal wastewater containing environmental concentrations of estrone, 17beta-estradiol, estriol, and 17alpha-ethinylestradiol were treated. Liquid chromatography-mass spectrometry analysis demonstrated that the studied steroid estrogens were completely oxidized in the wastewater reaction mixture after a 1-h treatment with either HRP (8-10 U ml(-1)) or laccase (20 U ml(-1)). Using the recombinant yeast assay, it was also confirmed that both enzymatic treatments were very efficient in removing the estrogenic activity of the studied steroid estrogens. The laccase-catalyzed process seemed to present great advantages over the HRP-catalyzed system for up-scale applications for the treatment of municipal wastewater.