Machine learning algorithms for spatial data. Case studies: Environmental pollution, natural hazards, renewable resources

This paper presents general problems and approaches for the spatial data analysis using machine learning algorithms. Machine learning is a very powerful approach to adaptive data analysis, modelling and visualisation. The key feature of the machine learning algorithms is that they learn from empirical data and can be used in cases when the modelled environmental phenomena are hidden, nonlinear, noisy and highly variable in space and in time. Most of the machines learning algorithms are universal and adaptive modelling tools developed to solve basic problems of learning from data: classification/pattern recognition, regression/mapping and probability density modelling. In the present report some of the widely used machine learning algorithms, namely artificial neural networks (ANN) of different architectures and Support Vector Machines (SVM), are adapted to the problems of the analysis and modelling of geo-spatial data. Machine learning algorithms have an important advantage over traditional models of spatial statistics when problems are considered in a high dimensional geo-feature spaces, when the dimension of space exceeds 5. Such features are usually generated, for example, from digital elevation models, remote sensing images, etc. An important extension of models concerns considering of real space constrains like geomorphology, networks, and other natural structures. Recent developments in semi-supervised learning can improve modelling of environmental phenomena taking into account on geo-manifolds. An important part of the study deals with the analysis of relevant variables and models' inputs. This problem is approached by using different feature selection/feature extraction nonlinear tools. To demonstrate the application of machine learning algorithms several interesting case studies are considered: digital soil mapping using SVM, automatic mapping of soil and water system pollution using ANN; natural hazards risk analysis (avalanches, landslides), assessments of renewable resources (wind fields) with SVM and ANN models, etc. The dimensionality of spaces considered varies from 2 to more than 30. Figures 1, 2, 3 demonstrate some results of the studies and their outputs. Finally, the results of environmental mapping are discussed and compared with traditional models of geostatistics.

Involvement of environmental mercury and lead in the etiology of neurodegenerative diseases.

The incidence of neurodegenerative disease like Parkinson's disease and Alzheimer's disease (AD) increases dramatically with age; only a small percentage is directly related to familial forms. The etiology of the most abundant, sporadic forms is complex and multifactorial, involving both genetic and environmental factors. Several environmental pollutants have been associated with neurodegenerative disorders. The present article focuses on results obtained in experimental neurotoxicology studies that indicate a potential pathogenic role of lead and mercury in the development of neurodegenerative diseases. Both heavy metals have been shown to interfere with a multitude of intracellular targets, thereby contributing to several pathogenic processes typical of neurodegenerative disorders, including mitochondrial dysfunction, oxidative stress, deregulation of protein turnover, and brain inflammation. Exposure to heavy metals early in development can precondition the brain for developing a neurodegenerative disease later in life. Alternatively, heavy metals can exert their adverse effects through acute neurotoxicity or through slow accumulation during prolonged periods of life. The pro-oxidant effects of heavy metals can exacerbate the age-related increase in oxidative stress that is related to the decline of the antioxidant defense systems. Brain inflammatory reactions also generate oxidative stress. Chronic inflammation can contribute to the formation of the senile plaques that are typical for AD. In accord with this view, nonsteroidal anti-inflammatory drugs and antioxidants suppress early pathogenic processes leading to Alzheimer's disease, thus decreasing the risk of developing the disease. The effects of lead and mercury were also tested in aggregating brain-cell cultures of fetal rat telencephalon, a three-dimensional brain-cell culture system. The continuous application for 10 to 50 days of non-cytotoxic concentrations of heavy metals resulted in their accumulation in brain cells and the occurrence of delayed toxic effects. When applied at non-toxic concentrations, methylmercury, the most common environmental form of mercury, becomes neurotoxic under pro-oxidant conditions. Furthermore, lead and mercury induce glial cell reactivity, a hallmark of brain inflammation. Both mercury and lead increase the expression of the amyloid precursor protein; mercury also stimulates the formation of insoluble beta-amyloid, which plays a crucial role in the pathogenesis of AD and causes oxidative stress and neurotoxicity in vitro. Taken together, a considerable body of evidence suggests that the heavy metals lead and mercury contribute to the etiology of neurodegenerative diseases and emphasizes the importance of taking preventive measures in this regard.

Environmental and biological controls on the geochemistry (δ18O, δ13C, Mg/Ca, and Sr/Ca) of living ostracods from Lake Geneva

ABSTRACT : Ostracods are benthic microcrustaceans enclosed in low-Mg calcite bivalves. Stable isotope compositions, Mg/Ca, and Sr/Ca ratios of ostracod fossil valves have proven useful to reconstruct past environmental conditions. Yet, several discrepancies persist and the influence of many factors remains unclear. It is the aim of this study to improve the use of ostracod valve geochemistry as palaeoenvironmental proxies by examining the extent of isotope fractionation and trace element partitioning during valve calcification. To achieve this, the environmental parameters (pH, temperature) and chemical composition of water (C-and O-isotope composition and calcium, magnesium, and strontium content) were measured at sites where living ostracods were sampled. The sampling was on a monthly basis over the course of one year at five different water depths (2, 5, 13, 33, and 70 m) in Lake Geneva, Switzerland. The one-year sampling enabled collection of environmental data for bottom and interstitial pore water. In littoral to sublittoral zones, C-isotope composition of DIC and the Mg/Ca and Sr/Ca ratios of water are found to vary concomitantly with water temperature. This is due to the precipitation of calcite, which is induced by higher photosynthetic activity as temperature and/or solar radiation intensify in summer. In deeper zones, environmental parameters remain largely constant throughout the year. Variations of pH, DIC concentrations and C-isotope compositions in interstitial water result from aerobic as well as anaerobic respiration, calcite dissolution and methanogenesis. Bathymetric distribution, life cycles, and habitats were derived for 15 ostracod species and are predominantly related to water temperature and sediment texture. O-isotope compositions of ostracod valves in Lake Geneva reflect that of water and temperature. However, offsets of up to 3 permil are observed in comparison with proposed inorganic calcite precipitation equilibrium composition. Deprotonation of HCO3- and/or salt effect at crystallisation sites may explain the disequilibrium observed for O-isotopic compositions. C-isotope compositions of ostracod valves are not as well constrained and appear to be controlled by a complex interaction between habitat preferences and seasonal as well as spatial variations of the DIC isotope composition. For infaunal forms, C-isotope compositions reflect mainly the variation of DIC isotope composition in interstitial pore waters. For epifaunal forms, C-isotope compositions reflect the seasonal variation of DIC isotope compositions. C-isotope compositions of ostracod valves is at equilibrium with DIC except for a small number of species (L. inopinata, L. sanctipatricii and possibly C. ophtalmica, and I. beauchampi). Trace element uptake differs considerably from species to species. For most epifaunal forms, trace element content follows the seasonal cycle, recording temperature increases and/or variations of Mg/Ca and Sr/Ca ratios of water. In contrast, infaunal forms are predominantly related to sediment pore water chemistry. RÉSUMÉ EN FRANÇAIS : Les ostracodes sont de petits crustacés benthiques qui possèdent une coquille faite de calcite à faible teneur en magnésium. La composition isotopique et les rapports Mg/Ca et Sr/Ca d'ostracodes fossiles ont été utilisés maintes fois avec succès pour effectuer des reconstructions paléoenvironnementales. Néanmoins, certains désaccords persistent sur l'interprétation de ces données. De plus, l'influence de certains facteurs pouvant biaiser le signal reste encore inconnue. Ainsi, le but de cette étude est de rendre plus performant l'emploi de la composition géochimique des ostracodes comme indicateur paléoenvironnemental. Pour réaliser cela, cinq sites situés dans le Léman à 2, 5, 13, 33 et 70 m de profondeur ont été choisis pour effectuer les échantillonnages. Chaque site a été visité une fois par mois durant une année. Les différents paramètres environnementaux (pH, température) ainsi que la composition géochimique de l'eau (composition isotopique de l'oxygène et du carbone ainsi que teneur en calcium, magnésium et strontium) ont été déterminés pour chaque campagne. Des ostracodes vivants ont été récoltés au cinq sites en même temps que les échantillons d'eau. Ce travail de terrain a permis de caractériser la géochimie de l'eau se trouvant juste au-dessus des sédiments ainsi que celle de l'eau se trouvant dans les interstices du sédiment. Dans les zones littorales à sublittorales, la composition isotopique du carbone inorganique dissout (CID) ainsi que les rapports Mg/Ca et Sr/Ca de l'eau varient linéairement avec la température. Ceci peut être expliqué par la précipitation de calcite qui est contrôlée par l'activité photosynthétique, variant elle même linéairement avec la température. Dans les zones plus profondes, les paramètres environnementaux restent relativement constants tout au long de l'année. Les variations du pH, de la concentration et de la composition isotopique du CID dans les sédiments résultent de la libération de carbone engendrée par la dégradation de la matière organique avec présence d'oxygène ou via réduction de nitrates et de sulfates, par la dissolution de carbonates, ainsi que par la méthanogenèse. La distribution bathymétrique, le cycle de vie ainsi que l'habitat de 15 espèces ont été déterminés. Ceux-ci sont principalement reliés à la température de l'eau et à la texture des sédiments. La composition isotopique de l'oxygène des valves d'ostracodes reflète celle de l'eau et la température qui régnait lors de la calcification. Néanmoins, des écarts pouvant aller jusqu'à 3 0/00 par rapport à l'équilibre théorique ont été obtenus. La déprotonation de HCO3 ou un 'effet de sel' pourrait être à l'origine du déséquilibre observé. La composition isotopique du carbone des valves d'ostracodes n'est pas aussi bien cernée. Celle-ci semble être principalement contrôlée par une interaction complexe entre l'habitat des ostracodes et les variations saisonnières et spatiales de la composition isotopique du CID. Pour les espèces endofaunes, la composition isotopique du carbone reflète principalement la variation de la composition isotopique du CID à l'intérieur des sédiments. Pour les formes épifaunes, c'est la variation saisonnière de la composition du CID qui contrôle celle de la coquille des ostracodes. En général, la composition isotopique du carbone des valves d'ostracodes est en équilibre avec celle de CID, hormis pour quelques rares espèces (L. inopinata, L. sanctipatricii et peut-être C. ophtalmica et I. beauchampi). L'incorporation des éléments traces diffère passablement d'une espèce à l'autre. Pour la plupart des espèces épifaunes, la teneur en éléments traces des coquilles reflète les variations saisonnières. Ces espèces semblent enregistrer les variations soit de la température soit des rapports Mg/Ca et Sr/Ca de l'eau. La teneur en élément traces des formes infaunales, au contraire, est principalement reliée à la chimie de l'eau interstitielle. RÉSUMÉ GRAND-PUBLIC : La connaissance de l'évolution du climat dans le futur est primordiale pour notre société, car elle permet de développer différentes stratégies pour faire face aux problèmes engendrés pas le changement climatique : stratégies environnementale, humanitaire, ou encore économique. Cette problématique est actuellement, à juste titre, sujet d'une vive préoccupation. La géologie peut-elle contribuer à l'effort communautaire entrepris? Naturellement, ce sont les climatologues qui sont sur le devant de la scène. Il n'empêche que ces derniers, pour pouvoir prédire l'avenir, doivent s'appuyer sur le passé. La géologie est alors d'un grand intérêt car c'est effectivement la seule science qui permette d'estimer les variations climatiques à grande échelle sur de longues périodes. Ainsi, voulant moi-même contribuer aux recherches menées dans ce domaine, je me suis tourné à la fin de mes études vers la paléoclimatologie, science qui a pour but de reconstruire le climat des temps anciens. Nous nous sommes rendu compte que l'évolution climatique de la région où nous habitons n'avait pas encore fait le sujet d'études approfondies. Il est pourtant important de connaître la variation locale des changements climatiques pour obtenir des modèles climatiques fiables. En conséquence, un vaste projet a vu le jour : reconstruire, à l'aide des sédiments du lac Léman, les variations paléoclimatiques et paléo-environnementales depuis le retrait du Glacier de Rhône, il y a environ 15'000 ans, jusqu'à nos jours. Pour ce genre de travail, la géochimie, qui est une forme de chimie, utilisée en science de la terre regroupant la chimie classique et la chimie isotopique, est une alliée particulièrement efficace. Elle permet en effet, via différentes mesures faites sur des archives géologiques (par exemple des fossiles ou des sédiments) d'obtenir des informations, souvent quantitatives, sur les conditions (le climat, la flore ou encore la bio productivité, etc...) qui régnaient il y a fort longtemps. Les coquilles d'ostracodes, qui sont de petits animaux vivant au fond des lacs, sont une des archives les plus prometteuses. Ces animaux sont des petits crustacés s'entourant d'une coquille calcaire qu'ils sécrètent eux-mêmes. A la mort de l'animal, la coquille est intégrée dans les sédiments et reste intacte à travers les âges. Des études ont montré qu'en analysant la géochimie de ces coquilles fossiles, il est possible de reconstruire les conditions environnementales qui régnaient à l'époque de vie de ces fossiles. Cette démarche nécessite qu'une condition bien précise soit remplie: la composition géochimique de la coquille doit enregistrer de manière fidèle la chimie de l'eau et/ou la température de l'eau présentes au moment de la sécrétion de la coquille. Le but spécifique de notre recherche a précisément été d'étudier la façon dont la chimie de l'eau ainsi que sa température sont enregistrées dans la coquillé des ostracodes. Une fois les relations entre ces divers paramètres dans l'étant actuel du système établies, il sera alors possible de les utiliser pour interpréter des données issues de coquilles fossiles. Pour ce faire, nous avons mesuré la température de l'eau de manière continue et récolté mensuellement des échantillons d'eau et des ostracodes vivants pendant une année. Cinq sites situés à 2, 5, 13, 33 et 70 mètres de profondeur ont été choisis pour effectuer ces échantillonnages dans le Léman. Le travail de terrain nous a amené à étudier la biologie de 15 espèces. Nous avons pu établir la profondeur à laquelle vivent ces animaux, leur période de développement ainsi que leur habitat respectifs. Ces résultats ont permis de mieux cerner la relation qu'il existe entre la chimie de l'eau, sa température et la composition géochimique des coquilles d'ostracodes. Nous avons ainsi pu confirmer que les coquilles d'ostracodes enregistrent de manière fidèle la composition chimique et isotopique de l'eau. De même, nous avons pu établir de manière plus précise l'effet de la température sur la géochimie des coquilles. Néanmoins, les relations trouvées entre ces trois éléments sont plus complexes pour certaines espèces, cette complexité étant souvent liée à un caractère spécifique de leur écologie. Nous avons mis en lumière certains effets qui biaisent les résultats et défini précisément les conditions dans lesquelles on peut s'attendre à avoir des difficultés dans leur interprétation. Maintenant que nous avons établi les relations entre le climat actuel et la composition géochimique des coquilles d'ostracodes actuels, nous pouvons, sur la base de ce modèle, reconstruire le climat depuis le retrait du Glacier du Rhône jusqu'à nos jours à l'aide d'ostracodes fossiles. Mais cela est une autre histoire et fera, je l'espère, le sujet de nos futures recherches.

NanoImpactNet - building a multi-stakeholder dialogue on the health and environmental impact of nanomaterials

NanoImpactNet (NIN) is a multidisciplinary European Commission funded network on the environmental, health and safety (EHS) impact of nanomaterials. The 24 founding scientific institutes are leading European research groups active in the fields of nanosafety, nanorisk assessment and nanotoxicology. This 4-year project is the new focal point for information exchange within the research community. Contact with other stakeholders is vital and their needs are being surveyed. NIN is communicating with 100s of stakeholders: businesses; internet platforms; industry associations; regulators; policy makers; national ministries; international agencies; standard-setting bodies and NGOs concerned by labour rights, EHS or animal welfare. To improve this communication, internet research, a questionnaire distributed via partners and targeted phone calls were used to identify stakeholders' interests and needs. Knowledge gaps and the necessity for further data mentioned by representatives of all stakeholder groups in the targeted phone calls concerned: • the potential toxic and safety hazards of nanomaterials throughout their lifecycles; • the fate and persistence of nanoparticles in humans, animals and the environment; • the associated risks of nanoparticle exposure; • greater participation in: the preparation of nomenclature, standards, methodologies, protocols and benchmarks; • the development of best practice guidelines; • voluntary schemes on responsibility; • databases of materials, research topics and themes, but also of expertise. These findings suggested that stakeholders and NIN researchers share very similar knowledge needs, and that open communication and free movement of knowledge will benefit both researchers and industry. Subsequently a workshop was organised by NIN focused on building a sustainable multi-stakeholder dialogue. Specific questions were asked to different stakeholder groups to encourage discussions and open communication. 1. What information do stakeholders need from researchers and why? The discussions about this question confirmed the needs identified in the targeted phone calls. 2. How to communicate information? While it was agreed that reporting should be enhanced, commercial confidentiality and economic competition were identified as major obstacles. It was recognised that expertise was needed in the areas of commercial law and economics for a wellinformed treatment of this communication issue. 3. Can engineered nanomaterials be used safely? The idea that nanomaterials are probably safe because some of them have been produced 'for a long time', was questioned, since many materials in common use have been proved to be unsafe. The question of safety is also about whether the public has confidence. New legislation like REACH could help with this issue. Hazards do not materialise if exposure can be avoided or at least significantly reduced. Thus, there is a need for information on what can be regarded as acceptable levels of exposure. Finally, it was noted that there is no such thing as a perfectly safe material but only boundaries. At this moment we do not know where these boundaries lie. The matter of labelling of products containing nanomaterials was raised, as in the public mind safety and labelling are connected. This may need to be addressed since the issue of nanomaterials in food, drink and food packaging may be the first safety issue to attract public and media attention, and this may have an impact on 'nanotechnology as a whole. 4. Do we need more or other regulation? Any decision making process should accommodate the changing level of uncertainty. To address the uncertainties, adaptations of frameworks such as REACH may be indicated for nanomaterials. Regulation is often needed even if voluntary measures are welcome because it mitigates the effects of competition between industries. Data cannot be collected on voluntary bases for example. NIN will continue with an active stakeholder dialogue to further build on interdisciplinary relationships towards a healthy future with nanotechnology.

The effect of a smoking ban introduction to environmental tobacco smoke exposure in Swiss hospitality workers : Abstracts of the 23rd Annual Conference of the International Society of Environmental Epidemiology (ISEE), September 13-16, 2011, Barcelona, Spain

Background and Aims: To protect the population from environmental tobacco smoke (ETS) Switzerland introduced a nationwide rather heterogeneous smoking ban in May 2010. The exposure situation of non-smoking hospitality workers before and after implementation of the new law is being assessed in a prospective cohort study. Methods: Exposure to ETS was measured using a novel method developed by the Institute for Work and Health in Lausanne. It is a passive sampler called MoNIC (Monitor of NICotine). The nicotine of the ETS is fixed onto a filter and transformed into salt of not volatile nicotine. Subsequently the number of passively smoked cigarettes is calculated. Badges were placed at the workplace as well as distributed to the participants for personal measuring. Additionally a salivary sample was taken to determine nicotine concentration. Results: At baseline Spearman's correlation between workplace and personal badge was 0.47. The average cigarette equivalents per day at the workplace obtained by badge significantly dropped from 5.1 (95%- CI: 2.4 to 7.9) at baseline to 0.3 (0.2 to 0.4) at first follow-up (n=29) three months later (p<0.001). For personal badges the number of passively smoked cigarettes declined from 1.5 (2.7 to 0.4) per day to 0.5 (0.3 to 0.8) (n=16).Salivary nicotine concentration in a subset of 13 participants who had worked on the day prior to the examination was 2.63 ng/ml before and 1.53 ng/ml after the ban (p=0.04). Spearman's correlation of salivary nicotine was 0.56 with workplace badge and 0.79 with personal badge concentrations. Conclusions: Workplace measurements clearly reflect the smoking regulation in a venue. The MoNIC badge proves to be a sensitive measuring device to determine personal ETS exposure and it is a demonstrative measure for communication with lay audiences and study participants as the number of passively smoked cigarettes is an easily conceivable result.

Bacterial biosensors for measuring availability of environmental pollutants

Abstract: Traditionally, pollution risk assessment is based on the measurement of a pollutant's total concentration in a sample. The toxicity of a given pollutant in the environment, however, is tightly linked to its bioavailability, which may differ significantly from the total amount. Physico-chemical and biological parameters strongly influence pollutant fate in terms of leaching, sequestration and biodegradation. Bacterial sensorreporters, which consist of living micro-organisms genetically engineered to produce specific output in response to target chemicals, offer an interesting alternative to monitoring approaches. Bacterial sensor-reporters detect bioavailable and/or bioaccessible compound fractions in samples. Currently, a variety of environmental pollutants can be targeted by specific biosensor-reporters. Although most of such strains are still confined to the lab, several recent reports have demonstrated utility of bacterial sensing-reporting in the field, with method detection limits in the nanomolar range. This review illustrates the general design principles for bacterial sensor-reporters, presents an overview of the existing biosensor-reporter strains with emphasis on organic compound detection. A specific focus throughout is on the concepts of bioavailability and bioaccessibility, and how bacteria-based sensing-reporting systems can help to improve our basic understanding of the different processes at work.

A framework for assessing the scale of influence of environmental factors on ecological patterns

The distribution of living organisms, habitats and ecosystems is primarily driven by abiotic environmental factors that are spatially structured. Assessing the spatial structure of environmental factors, e.g., through spatial autocorrelation analyses (SAC), can thus help us understand their scale of influence on the distribution of organisms, habitats, and ecosystems. Yet SAC analyses of environmental factors are still rarely performed in biogeographic studies. Here, we describe a novel framework that combines SAC and statistical clustering to identify scales of spatial patterning of environmental factors, which can then be interpreted as the scales at which those factors influence the geographic distribution of biological and ecological features. We illustrate this new framework with datasets at different spatial or thematic resolutions. This framework is conceptually and statistically robust, providing a valuable approach to tackle a wide range of issues in ecological and environmental research and particularly when building predictors for ecological models. The new framework can significantly promote fundamental research on all spatially-structured ecological patterns. It can also foster research and application in such fields as global change ecology, conservation planning, and landscape management.