Multitask learning of environmental spatial data

The present research deals with an application of artificial neural networks for multitask learning from spatial environmental data. The real case study (sediments contamination of Geneva Lake) consists of 8 pollutants. There are different relationships between these variables, from linear correlations to strong nonlinear dependencies. The main idea is to construct a subsets of pollutants which can be efficiently modeled together within the multitask framework. The proposed two-step approach is based on: 1) the criterion of nonlinear predictability of each variable ?k? by analyzing all possible models composed from the rest of the variables by using a General Regression Neural Network (GRNN) as a model; 2) a multitask learning of the best model using multilayer perceptron and spatial predictions. The results of the study are analyzed using both machine learning and geostatistical tools.

Growth decrease in Alpine whitefish: investigating the relative contribution of fishing induced selection and environmental change and its implication for conservation measures

La pression exercée par les activités humaines menace pratiquement tous les écosystèmes aquatiques du globe. Ainsi, sous l'effet de divers facteurs tels que la pollution, le réchauffement climatique ou encore la pêche industrielle, de nombreuses populations de poissons ont vu leurs effectifs chuter et divers changements morphologiques ont été observés. Dans cette étude, nous nous sommes intéressés à une menace particulière: la sélection induite par la pêche sur la croissance des poissons. En effet, la génétique des populations prédit que la soustraction régulière des individus les plus gros peut entraîner des modifications rapides de certains traits physiques comme la croissance individuelle. Cela a par ailleurs été observé dans de nombreuses populations marines ou lacustres, dont les populations de féras, bondelles et autres corégones des lacs suisses. Toutefois, malgré un nombre croissant d'études décrivant ce phénomène, peu de plans de gestion en tiennent compte, car l'importance des effets génétiques liés à la pêche est le plus souvent négligée par rapport à l'impact des changements environnementaux. Le but premier de cette étude a donc été de quantifier l'importance des facteurs génétiques et environnementaux. Dans le premier chapitre, nous avons étudié la population de palée du lac de Joux (Coregonus palaea). Nous avons déterminé les différentiels de sélection dus à la pêche, c'est-à-dire l'intensité de la sélection sur le taux de croissance, ainsi que les changements nets de croissance au cours du temps. Nous avons observé une baisse marquée de croissance et un différentiel de sélection important indiquant qu'au moins 30% de la diminution de croissance observée était due à la pression de sélection induite par la pêche. Dans le deuxième chapitre, nous avons effectué les mêmes analyses sur deux espèces proches du lac de Brienz (C. albellus et C. fatioi) et avons observé des effets similaires dont l'intensité était spécifique à chaque espèce. Dans le troisième chapitre, nous avons analysé deux autres espèces : C. palaea et C. confusus du lac de Bienne, et avons constaté que le lien entre la pression de sélection et la diminution de croissance était influencé par des facteurs environnementaux. Finalement, dans le dernier chapitre, nous avons étudié les effets potentiels de différentes modifications de la taille des mailles des filets utilisés pour la pêche à l'aide de modèles mathématiques. Nous concluons que la pêche a un effet génétique non négligeable (et donc peu réversible) sur la croissance individuelle dans les populations observée, que cet effet est lié à la compétition pour la nourriture et à la qualité de l'environnement, et que certaines modifications simples de la taille des mailles des filets de pêche pourraient nettement diminuer l'effet de sélection et ainsi ralentir, voir même renverser la diminution de croissance observée.

HIV disclosure and nondisclosure among migrant women from sub-Saharan Africa living in Switzerland.

No study to date has focused specifically on the reasons for and against disclosure of HIV-positive status among sub-Saharan migrant women. Thirty HIV-positive women from 11 sub-Saharan countries living in French-speaking Switzerland participated in semi-structured individual interviews. The reasons women reported for disclosure or nondisclosure of their HIV serostatus were classified into three categories: social, medical, and ethical. The women identified the stigma associated with HIV as a major social reason for nondisclosure. However, this study identifies new trends related to disclosure for medical and ethical reasons. Being undetectable played an important role in the life of sub-Saharan migrant women, and analysis revealed their medical reasons for both disclosure and nondisclosure. Disclosure to new sexual partners occurred when women had a more positive perception about HIV and when they believed themselves to be in a long-term relationship. Women reported nondisclosure to family members when they did not need help outside the support provided by the medical and social fields. The results on ethical reasons suggested that challenging stigma was a reason for disclosure. Since the women' perceptions on HIV changed when they came to see it as a chronic disease, disclosure occurred in an attempt to normalize life with HIV in their communities in migration and to challenge racism and discrimination. Our findings can help health providers better understand the communication needs of sub-Saharan migrant women with respect to HIV/AIDS and sexuality and offer them adequate disclosure advice that takes into account migration and gender issues.

Simultaneous determination of Pu and Am radioisotopes in environmental samples: a tool for determining origin and release date

A radiochemical procedure was developed for the sequential determination of Pu and Am radioisotopes in environmental samples. The radioisotope activities were then used to assess the origin and release date of the environmental plutonium. The radioanalytical procedure is based on the separation of Pu and Am on selective extraction chromatographic resins (Eichrom TEVA and DGA). Alpha sources were prepared by electrodeposition on stainless steel discs, and the alpha emitting radionuclides (238Pu, 239,240Pu and 241Am) were measured by alpha spectrometry. For the determination of the beta emitting 241Pu, the Pu alpha source was leached in hot concentrated nitric acid and the Pu fraction further purified by extraction chromatography on a small column of TEVA resin (100 μg of resin in a pipette tip). 241Pu is then measured by ultra low level liquid scintillation counting. Due to the lack of reference material for 241Pu, the proposed radiochemical method was nevertheless validated using four IAEA reference sediments with information values of 241Pu. The proposed method was then used to determine the 238Pu, 239,240Pu, 241Pu and 241Am activity concentrations in alpine soils of France and Switzerland. The soil is the primary receptor of the atmospheric radioactive fallout and, because of the strong binding interaction with soils particles, the isotopes are little fractionated. Therefore, the activity ratios 241Pu/239+240Pu and 238Pu/239,240Pu in soil samples were used to determine the origin (source) and date of the Pu contamination in the investigated alpine sites. The 241Pu/239,240Pu and 238Pu/239,240Pu activity ratios confirmed that the main origin of Pu in the alpine soils was the global fallout from the nuclear bomb tests (NBT) in the fifties and sixties. Furthermore, the 241Pu/241Am activity ratios were used to determine the age of the Pu contamination, which is also an important data for distinguishing the Pu sources. The estimation of the date of the contamination, by the 241Pu/241Am age-dating method, further confirmed the NBT as the Pu source. However, the 241Pu/241Am dating method was limited to samples where Pu-Am fractionation was insignificant. If any, the contribution of the Chernobyl accident in the studied sites is negligible.

Accuracy of plant assemblage predictions from species distribution models varies along environmental gradients

Aim: Climatic niche modelling of species and community distributions implicitly assumes strong and constant climatic determinism across geographic space. This assumption had however never been tested so far. We tested it by assessing how stacked-species distribution models (S-SDMs) perform for predicting plant species assemblages along elevation. Location: Western Swiss Alps. Methods: Using robust presence-absence data, we first assessed the ability of topo-climatic S-SDMs to predict plant assemblages in a study area encompassing a 2800 m wide elevation gradient. We then assessed the relationships among several evaluation metrics and trait-based tests of community assembly rules. Results: The standard errors of individual SDMs decreased significantly towards higher elevations. Overall, the S-SDM overpredicted far more than they underpredicted richness and could not reproduce the humpback curve along elevation. Overprediction was greater at low and mid-range elevations in absolute values but greater at high elevations when standardised by the actual richness. Looking at species composition, the evaluation metrics accounting for both the presence and absence of species (overall prediction success and kappa) or focusing on correctly predicted absences (specificity) increased with increasing elevation, while the metrics focusing on correctly predicted presences (Jaccard index and sensitivity) decreased. The best overall evaluation - as driven by specificity - occurred at high elevation where species assemblages were shown to be under significant environmental filtering of small plants. In contrast, the decreased overall accuracy in the lowlands was associated with functional patterns representing any type of assembly rule (environmental filtering, limiting similarity or null assembly). Main Conclusions: Our study reveals interesting patterns of change in S-SDM errors with changes in assembly rules along elevation. Yet, significant levels of assemblage prediction errors occurred throughout the gradient, calling for further improvement of SDMs, e.g., by adding key environmental filters that act at fine scales and developing approaches to account for variations in the influence of predictors along environmental gradients.

Multiple Kernel Learning of Environmental Data. Case study: Analysis and Mapping of Wind Fields

The paper presents the Multiple Kernel Learning (MKL) approach as a modelling and data exploratory tool and applies it to the problem of wind speed mapping. Support Vector Regression (SVR) is used to predict spatial variations of the mean wind speed from terrain features (slopes, terrain curvature, directional derivatives) generated at different spatial scales. Multiple Kernel Learning is applied to learn kernels for individual features and thematic feature subsets, both in the context of feature selection and optimal parameters determination. An empirical study on real-life data confirms the usefulness of MKL as a tool that enhances the interpretability of data-driven models.

France's Charter for the Environment : of Presidents, Principles and Environmental Protection

France amended its constitution in 2005 to include a Charter for the Environment. The Charter lays out France's commitment to supporting the right to a 'balanced environment'. This article first traces the Charter's origins to a legacy-building presidential initiative. Jacques Chirac decided to invest in a neglected policy domain in which his own majority had shown little interest. He was obliged to intervene repeatedly in order to bring this project to a successful conclusion. In doing so, he staked out environmental affairs as an area of potential presidential supremacy. Next, the content of the Charter is examined. In this document, French traditions of universalism come together with an international movement for anticipatory environmental protection. This is reflected in the constitutionalisation of the precautionary principle, which emerged as the most controversial part of the Charter. The debates this provoked tended to caricature a risk-management principle whose meaning has been carefully refined to forestall objections. Finally, the Charter's potential efficacy is analysed. The post-Charter record of legislative and judicial activity concerning the environment is meagre, but not wholly inauspicious.

Environmental pollution promotes selection of microbial degradation pathways

Astonishing as it may seem, one organism's waste is often ideal food for another. Many waste products generated by human activities are routinely degraded by microorganisms under controlled conditions during waste-water treatment. Toxic pollutants resulting from inadvertent releases, such as oil spills, are also consumed by bacteria, the simplest organisms on Earth. Biodegradation of toxic or particularly persistent compounds, however, remains problematic. What has escaped the attention of many is that bacteria exposed to pollutants can adapt to them by mutating or acquiring degradative genes. These bacteria can proliferate in the environment as a result of the selection pressures created by pollutants. The positive outcome of selection pressure is that harmful compounds may eventually be broken down completely through biodegradation. The downside is that biodegradation may require extremely long periods of time. Although the adaptation process has been shown to be reproducible, it remains very difficult to predict.

Very high-resolution environmental predictors in species distribution models: moving beyond topography?

Recent advances in remote sensing technologies have facilitated the generation of very high resolution (VHR) environmental data. Exploratory studies suggested that, if used in species distribution models (SDMs), these data should enable modelling species' micro-habitats and allow improving predictions for fine-scale biodiversity management. In the present study, we tested the influence, in SDMs, of predictors derived from a VHR digital elevation model (DEM) by comparing the predictive power of models for 239 plant species and their assemblages fitted at six different resolutions in the Swiss Alps. We also tested whether changes of the model quality for a species is related to its functional and ecological characteristics. Refining the resolution only contributed to slight improvement of the models for more than half of the examined species, with the best results obtained at 5 m, but no significant improvement was observed, on average, across all species. Contrary to our expectations, we could not consistently correlate the changes in model performance with species characteristics such as vegetation height. Temperature, the most important variable in the SDMs across the different resolutions, did not contribute any substantial improvement. Our results suggest that improving resolution of topographic data only is not sufficient to improve SDM predictions - and therefore local management - compared to previously used resolutions (here 25 and 100 m). More effort should be dedicated now to conduct finer-scale in-situ environmental measurements (e.g. for temperature, moisture, snow) to obtain improved environmental measurements for fine-scale species mapping and management.

Whole-cell living biosensors--are they ready for environmental application?

Since the development of the first whole-cell living biosensor or bioreporter about 15 years ago, construction and testing of new genetically modified microorganisms for environmental sensing and reporting has proceeded at an ever increasing rate. One and a half decades appear as a reasonable time span for a new technology to reach the maturity needed for application and commercial success. It seems, however, that the research into cellular biosensors is still mostly in a proof-of-principle or demonstration phase and not close to extensive or commercial use outside of academia. In this review, we consider the motivations for bioreporter developments and discuss the suitability of extant bioreporters for the proposed applications to stimulate complementary research and to help researchers to develop realistic objectives. This includes the identification of some popular misconceptions about the qualities and shortcomings of bioreporters.