Quantitative analysis of crystal/grain sizes and their distributions in 2D and 3D

We review methods to estimate the average crystal (grain) size and the crystal (grain) size distribution in solid rocks. Average grain sizes often provide the base for stress estimates or rheological calculations requiring the quantification of grain sizes in a rock's microstructure. The primary data for grain size data are either 1D (i.e. line intercept methods), 2D (area analysis) or 3D (e.g., computed tomography, serial sectioning). These data have been used for different data treatments over the years, whereas several studies assume a certain probability function (e.g., logarithm, square root) to calculate statistical parameters as the mean, median, mode or the skewness of a crystal size distribution. The finally calculated average grain sizes have to be compatible between the different grain size estimation approaches in order to be properly applied, for example, in paleo-piezometers or grain size sensitive flow laws. Such compatibility is tested for different data treatments using one- and two-dimensional measurements. We propose an empirical conversion matrix for different datasets. These conversion factors provide the option to make different datasets compatible with each other, although the primary calculations were obtained in different ways. In order to present an average grain size, we propose to use the area-weighted and volume-weighted mean in the case of unimodal grain size distributions, respectively, for 2D and 3D measurements. The shape of the crystal size distribution is important for studies of nucleation and growth of minerals. The shape of the crystal size distribution of garnet populations is compared between different 2D and 3D measurements, which are serial sectioning and computed tomography. The comparison of different direct measured 3D data; stereological data and direct presented 20 data show the problems of the quality of the smallest grain sizes and the overestimation of small grain sizes in stereological tools, depending on the type of CSD. (C) 2011 Published by Elsevier Ltd.

Stationarity analysis of historical flood series in France and Spain (14th-20th centuries)

Interdisciplinary frameworks for studying natural hazards and their temporal trends have an important potential in data generation for risk assessment, land use planning, and therefore the sustainable management of resources. This paper focuses on the adjustments required because of the wide variety of scientific fields involved in the reconstruction and characterisation of flood events for the past 1000 years. The aim of this paper is to describe various methodological aspects of the study of flood events in their historical dimension, including the critical evaluation of old documentary and instrumental sources, flood-event classification and hydraulic modelling, and homogeneity and quality control tests. Standardized criteria for flood classification have been defined and applied to the Isère and Drac floods in France, from 1600 to 1950, and to the Ter, the Llobregat and the Segre floods, in Spain, from 1300 to 1980. The analysis on the Drac and Isère data series from 1600 to the present day showed that extraordinary and catastrophic floods were not distributed uniformly in time. However, the largest floods (general catastrophic floods) were homogeneously distributed in time within the period 1600¿1900. No major flood occurred during the 20th century in these rivers. From 1300 to the present day, no homogeneous behaviour was observed for extraordinary floods in the Spanish rivers. The largest floods were uniformly distributed in time within the period 1300-1900, for the Segre and Ter rivers.

Radar analysis of the life cycle of mesoscale convective systems during the 10 June 2000 event

The 10 June 2000 event was the largest flash flood event that occurred in the Northeast of Spain in the late 20th century, both as regards its meteorological features and its considerable social impact. This paper focuses on analysis of the structures that produced the heavy rainfalls, especially from the point of view of meteorological radar. Due to the fact that this case is a good example of a Mediterranean flash flood event, a final objective of this paper is to undertake a description of the evolution of the rainfall structure that would be sufficiently clear to be understood at an interdisciplinary forum. Then, it could be useful not only to improve conceptual meteorological models, but also for application in downscaling models. The main precipitation structure was a Mesoscale Convective System (MCS) that crossed the region and that developed as a consequence of the merging of two previous squall lines. The paper analyses the main meteorological features that led to the development and triggering of the heavy rainfalls, with special emphasis on the features of this MCS, its life cycle and its dynamic features. To this end, 2-D and 3-D algorithms were applied to the imagery recorded over the complete life cycle of the structures, which lasted approximately 18 h. Mesoscale and synoptic information were also considered. Results show that it was an NS-MCS, quasi-stationary during its stage of maturity as a consequence of the formation of a convective train, the different displacement directions of the 2-D structures and the 3-D structures, including the propagation of new cells, and the slow movement of the convergence line associated with the Mediterranean mesoscale low.

The application of trend surface models to the analysis of time factors in Swiss cancer mortality.

To study different temporal components on cancer mortality (age, period and cohort) methods of graphic representation were applied to Swiss mortality data from 1950 to 1984. Maps using continuous slopes ("contour maps") and based on eight tones of grey according to the absolute distribution of rates were used to represent the surfaces defined by the matrix of various age-specific rates. Further, progressively more complex regression surface equations were defined, on the basis of two independent variables (age/cohort) and a dependent one (each age-specific mortality rate). General patterns of trends in cancer mortality were thus identified, permitting definition of important cohort (e.g., upwards for lung and other tobacco-related neoplasms, or downwards for stomach) or period (e.g., downwards for intestines or thyroid cancers) effects, besides the major underlying age component. For most cancer sites, even the lower order (1st to 3rd) models utilised provided excellent fitting, allowing immediate identification of the residuals (e.g., high or low mortality points) as well as estimates of first-order interactions between the three factors, although the parameters of the main effects remained still undetermined. Thus, the method should be essentially used as summary guide to illustrate and understand the general patterns of age, period and cohort effects in (cancer) mortality, although they cannot conceptually solve the inherent problem of identifiability of the three components.

A comparative phenotypic and genomic analysis of C57BL/6J and C57BL/6N mouse strains.

BACKGROUND: The mouse inbred line C57BL/6J is widely used in mouse genetics and its genome has been incorporated into many genetic reference populations. More recently large initiatives such as the International Knockout Mouse Consortium (IKMC) are using the C57BL/6N mouse strain to generate null alleles for all mouse genes. Hence both strains are now widely used in mouse genetics studies. Here we perform a comprehensive genomic and phenotypic analysis of the two strains to identify differences that may influence their underlying genetic mechanisms. RESULTS: We undertake genome sequence comparisons of C57BL/6J and C57BL/6N to identify SNPs, indels and structural variants, with a focus on identifying all coding variants. We annotate 34 SNPs and 2 indels that distinguish C57BL/6J and C57BL/6N coding sequences, as well as 15 structural variants that overlap a gene. In parallel we assess the comparative phenotypes of the two inbred lines utilizing the EMPReSSslim phenotyping pipeline, a broad based assessment encompassing diverse biological systems. We perform additional secondary phenotyping assessments to explore other phenotype domains and to elaborate phenotype differences identified in the primary assessment. We uncover significant phenotypic differences between the two lines, replicated across multiple centers, in a number of physiological, biochemical and behavioral systems. CONCLUSIONS: Comparison of C57BL/6J and C57BL/6N demonstrates a range of phenotypic differences that have the potential to impact upon penetrance and expressivity of mutational effects in these strains. Moreover, the sequence variants we identify provide a set of candidate genes for the phenotypic differences observed between the two strains.

Therapeutic drug monitoring and metabolites profil analysis of antiviral drugs

The widespread use of combination antiretroviral therapy (ARVs) has considerably improved the prognosis of patients infected with HIV. Conversely, considerable advances have been recently realized for the therapy of hepatitis C infection with the recent advent of potent new anti-HCV drugs that allow an increasing rate HCV infection cure. Despite their overall efficacy, a significant number of patients do not achieve or maintain adequate clinical response, defined as an undetectable viral load for HIV, and a sustained virological response (or cure) in HCV infection. Treatment failure therefore still remains an important issue besides drugs toxicities and viral resistance which is not uncommon in a significant percentage of patients who do not reach adequate virological suppression. The reasons of variability in drug response are multifactorial and apart from viral genetics, other factors such as environmental factors, drug- drug interactions, and imperfect compliance may have profound impact on antiviral drugs' clinical response. The possibility of measuring plasma concentration of antiviral drugs enables to guide antiviral drug therapy and ensure optimal drug exposure. The overall objective of this research was to widen up the current knowledge on pharmacokinetic and pharmacogenetic factors that influence the clinical response and toxicity of current and newly approved antiretroviral and anti-HCV drugs. To that endeavour, analytical methods using liquid chromatography coupled with tandem mass spectrometry have been developed and validated for the precise and accurate measurement of new antiretroviral and anti-HCV drugs . These assays have been applied for the TDM of ARVs and anti-HCV in patients infected with either HIV or HCV respectively, and co-infected with HIV- HCV. A pharmacokinetic population model was developed to characterize inter and intra-patient variability of rilpivirine, the latest marketed Non Nucleoside Reverse transcriptase (NNRTI) Inhibitor of HIVand to identify genetic and non genetic covariates influencing rilpivirine exposure. None of the factors investigated so far showed however any influence of RPV clearance. Importantly, we have found that the standard daily dosage regimen (25 mg QD) proposed for rilpivirine results in concentrations below the proposed therapeutic target in about 40% of patients. In these conditions, virologie escape is a potential risk that remains to be further investigated, notably via the TDM approach that can be a useful tool to identify patients who are at risk for being exposed to less than optimal levels of rilpivirine in plasma. Besides the last generation NNRTI rilpivirine, we have studied efavirenz, the major NNRTI clinically used so far. Namely for efavirenz, we aimed at identifying a potential new marker of toxicity that may be incriminated for the neuropsychological sides effects and hence discontinuation of efavirenz therapy. To that endeavour, a comprehensive analysis of phase I and phase II metabolites profiles has been performed in plasma, CSF and in urine from patients under efavirenz therapy. We have found that phase II metabolites of EFV constitute the major species circulating in blood, sometimes exceeding the levels of the parent drug efavirenz. Moreover we have identified a new metabolite of efavirenz in humans, namely the 8-OH-EFV- sulfate which is present at high concentrations in all body compartments from patients under efavirenz therapy. These investigations may open the way to possible alternate phenotypic markers of efavirenz toxicity. Finally, the specific influence of P-glycoprotein on the cellular disposition of a series ARVs (NNRTIs and Pis] has been studies in in vitro cell systems using the siRNA silencing approach. -- Depuis l'introduction de la thérapie antirétrovirale (ARVs) la morbidité et la mortalité liées au VIH ont considérablement diminué. En parallèle le traitement contre le virus de l'hépatite C (VHC) a connu récemment d'énormes progrès avec l'arrivée de nouveaux médicaments puissants, ce qui a permis une augmentation considérable de la guérison de l'infection par le VHC. En dépit de l'efficacité de ces traitements antiviraux, les échecs thérapeutiques ainsi que les effets secondaires des traitements restent un problème important. Une réponse imparfaite ou la toxicité du traitement est certainement multifactorielle. Le suivi thérapeutique des médicaments [Therapeutic Drug Monitoring TDM) à travers la mesure des concentrations plasmatiques constitue une approche importante pour guider le traitement médicamenteux et de s'assurer que les patients sont exposés à des concentrations optimales des médicaments dans le sang, et puissent tirer tout le bénéfice potentiel du traitement. L'objectif global de cette thèse était d'étudier les facteurs pharmacocinétiques et pharmacogénétiques qui influencent l'exposition des médicaments antiviraux (ARVs et anti- VHC) récemment approuvés. A cet effet, des méthodes de quantification des concentrations plasmatiques des médicaments antirétroviraux, anti-VHC ainsi que pour certains métabolites ont été développées et validées en utilisant la Chromatographie liquide couplée à la spectrométrie de masse tandem. Ces méthodes ont été utilisées pour le TDM des ARVs et pour les agents anti-VHC chez les patients infectés par le VIH, et le VHC, respectivement, mais aussi chez les patients co-infectés par le VIH-VHC. Un modèle de pharmacocinétique de population a été développé pour caractériser la variabilité inter-et intra-patient du médicament rilpivirine, un inhibiteur non nucléosidique de la transcriptase de VIH et d'identifier les variables génétiques et non génétiques influençant l'exposition au médicament. Aucun des facteurs étudiés n'a montré d'influence notable sur la clairance de la rilpivirine. Toutefois, la concentration résiduelle extrapolée selon le modèle de pharmacocinétique de population qui a été développé, a montré qu'une grande proportion des patients présente des concentrations minimales inférieures à la cible thérapeutique proposée. Dans ce contexte, la relation entre les concentrations minimales et l'échappement virologique nécessite une surveillance étroite des taux sanguins des patients recevant de la rilpivirine. A cet effet, le suivi thérapeutique est un outil important pour l'identification des patients à risque soient sous-exposés à lai rilpivirine. Pour identifier de nouveaux marqueurs de la toxicité qui pourraient induire l'arrêt du traitement, le profil des métabolites de phase I et de phase II a été étudié dans différentes matrices [plasma, LCR et urine) provenant de patients recevant de l'efavirenz. Les métabolites de phase II, qui n'avaient à ce jour jamais été investigués, constituent les principales espèces présentes dans les matrices étudiées. Au cours de ces investigations, un nouveau métabolite 8- OH-EFV-sulfate a été identifié chez l'homme, et ce dernier est. présent à des concentrations importantes. L'influence de certains facteurs pharmacogénétique des patients sur le profil des métabolites a été étudiée et ouvre la voie à de possibles nouveaux marqueurs phénotypiques alternatifs qui pourraient possiblement mieux prédire la toxicité associée au traitement par l'efavirenz. Finalement, nous nous sommes intéressés à étudier dans un modèle in vitro certains facteurs, comme la P-glycoprotéine, qui influencent la disposition cellulaire de certains médicaments antirétroviraux, en utilisant l'approche par la technologie du siRNA permettant de bloquer sélectivement l'expression du gène de cette protéine d'efflux des médicaments. -- Depuis l'introduction de la thérapie antiretrovirale (ARVs] la morbidité et la mortalité liées au VIH ont considérablement diminué. En parallèle le traitement contre le virus de l'hépatite C (VHC) a connu récemment d'énormes progrès avec l'arrivée de nouveaux médicaments puissants, ce qui a permis une augmentation considérable de la guérison de l'infection par le VHC. En dépit de l'efficacité de ces traitements antiviraux, les échecs thérapeutiques ainsi que les effets secondaires des traitements restent un problème important. Il a pu être démontré que la concentration de médicament présente dans l'organisme est corrélée avec l'efficacité clinique pour la plupart des médicaments agissant contre le VIH et contre le VHC. Les médicaments antiviraux sont généralement donnés à une posologie fixe et standardisée, à tous les patients, il existe cependant une importante variabilité entre les concentrations sanguines mesurées chez les individus. Cette variabilité peut être expliquée par plusieurs facteurs démographiques, environnementaux ou génétiques. Dans ce contexte, le suivi des concentrations sanguines (ou Therapeutic Drug Monitoring, TDM) permet de contrôler que les patients soient exposés à des concentrations suffisantes (pour bloquer la réplication du virus dans l'organisme) et éviter des concentrations excessives, ce qui peut entraîner l'apparition d'intolérence au traitement. Le but de ce travail de thèse est d'améliorer la compréhension des facteurs pharmacologiques et génétiques qui peuvent influencer l'efficacité et/ou la toxicité des médicaments antiviraux, dans le but d'améliorer le suivi des patients. A cet effet, des méthodes de dosage très sensibles et ont été mises au point pour permettre de quantifier les médicaments antiviraux dans le sang et dans d'autres liquides biologiques. Ces méthodes de dosage sont maintenant utilisées d'une part dans le cadre de la prise en charge des patients en routine et d'autre part pour diverses études cliniques chez les patients infectés soit par le HIV, le HCV ou bien coinfectés par les deux virus. Une partie de ce travail a été consacrée à l'investigation des différents facteurs démographiques, génétiques et environnementaux qui pourraient l'influencer la réponse clinique à la rilpivirine, un nouveau médicament contre le VIH. Toutefois, parmi tous les facteurs étudiés à ce jour, aucun n'a permis d'expliquer la variabilité de l'exposition à la rilpivirine chez les patients. On a pu cependant observer qu'à la posologie standard recommandée, un pourcentage relativement élevé de patients pourrait présenter des concentrations inférieures à la concentration sanguine minimale actuellement proposée. Il est donc utile de surveiller étroitement les concentrations de rilpivirine chez les patients pour identifier sans délai ceux qui risquent d'être sous-exposés. Dans l'organisme, le médicament subit diverses transformations (métabolisme) par des enzymes, notamment dans le foie, il est transporté dans les cellules et tissus par des protéines qui modulent sa concentration au site de son action pharmacologique. A cet effet, différents composés (métabolites) produits dans l'organisme après l'administration d'efavirenz, un autre médicament anti-VIH, ont été étudiés. En conclusion, nous nous sommes intéressés à la fois aux facteurs pharmacologiques et génétiques des traitements antiviraux, une approche qui s'inscrit dans l'optique d'une stratégie globale de prise en charge du patient. Dans ce contexte, le suivi des concentrations sanguines de médicaments constitue une des facettes du domaine émergent de la Médecine Personnalisée qui vise à maximiser le bénéfice thérapeutique et le profil de tolérance des médicaments antiviraux

Numerical analysis of the impact of charcoal production on soil hydrological behavior, runoff response and erosion susceptibility

The impact of charcoal production on soil hydraulic properties, runoff response and erosion susceptibility were studied in both field and simulation experiments. Core and composite samples, from 12 randomly selected sites within the catchment of Kotokosu were taken from the 0-10 cm layer of a charcoal site soil (CSS) and adjacent field soils (AFS). These samples were used to determine saturated hydraulic conductivity (Ksat), bulk density, total porosity, soil texture and color. Infiltration, surface albedo and soil surface temperature were also measured in both CSS and AFS. Measured properties were used as entries in a rainfall runoff simulation experiment on a smooth (5 % slope) plot of 25 x 25 m grids with 10 cm resolutions. Typical rainfall intensities of the study watershed (high, moderate and low) were applied to five different combinations of Ks distributions that could be expected in this landscape. The results showed significantly (p < 0.01) higher flow characteristics of the soil under charcoal kilns (increase of 88 %). Infiltration was enhanced and runoff volume reduced significantly. The results showed runoff reduction of about 37 and 18 %, and runoff coefficient ranging from 0.47-0.75 and 0.04-0.39 or simulation based on high (200 mm h-1) and moderate (100 mm h-1) rainfall events over the CSS and AFS areas, respectively. Other potential impacts of charcoal production on watershed hydrology were described. The results presented, together with watershed measurements, when available, are expected to enhance understanding of the hydrological responses of ecosystems to indiscriminate charcoal production and related activities in this region.

Molecular analysis of sleep.

Rest or sleep in all animal species constitutes a period of quiescence necessary for recovery from activity. Whether rest and activity observed in all organisms share a similar fundamental molecular basis with sleep and wakefulness in mammals has not yet been established. In addition and in contrast to the circadian system, strong evidence that sleep is regulated at the transcriptional level is lacking. Nevertheless, several studies indicate that single genesmay regulate some specific aspects of sleep. Efforts to better understand or confirm the role of known neurotransmission pathways in sleep-wake regulation using transgenic approaches resulted so far in only limited new insights. Recent gene expression profiling efforts in rats, mice, and fruit flies are promising and suggest that only a few gene categories are differentially regulated by behavioral state. How molecular analysis can help us to understand sleep is the focus of this chapter.

Quality coding by neural populations in the early olfactory pathway: analysis using information theory

In this article, we analyze the ability of the early olfactory system to detect and discriminate different odors by means of information theory measurements applied to olfactory bulb activity images. We have studied the role that the diversity and number of receptor neuron types play in encoding chemical information. Our results show that the olfactory receptors of the biological system are low correlated and present good coverage of the input space. The coding capacity of ensembles of olfactory receptors with the same receptive range is maximized when the receptors cover half of the odor input space - a configuration that corresponds to receptors that are not particularly selective. However, the ensemble's performance slightly increases when mixing uncorrelated receptors of different receptive ranges. Our results confirm that the low correlation between sensors could be more significant than the sensor selectivity for general purpose chemo-sensory systems, whether these are biological or biomimetic.

Pharmacokinetic and pharmacodynamic analysis of efavirenz dose reduction using an in vitro-in vivo extrapolation model.

The pharmacokinetics (PK) of efavirenz (EFV) is characterized by marked interpatient variability that correlates with its pharmacodynamics (PD). In vitro-in vivo extrapolation (IVIVE) is a "bottom-up" approach that combines drug data with system information to predict PK and PD. The aim of this study was to simulate EFV PK and PD after dose reductions. At the standard dose, the simulated probability was 80% for viral suppression and 28% for central nervous system (CNS) toxicity. After a dose reduction to 400 mg, the probabilities of viral suppression were reduced to 69, 75, and 82%, and those of CNS toxicity were 21, 24, and 29% for the 516 GG, 516 GT, and 516 TT genotypes, respectively. With reduction of the dose to 200 mg, the probabilities of viral suppression decreased to 54, 62, and 72% and those of CNS toxicity decreased to 13, 18, and 20% for the 516 GG, 516 GT, and 516 TT genotypes, respectively. These findings indicate how dose reductions might be applied in patients with favorable genetic characteristics.

A time-domain method for the analysis of thermal impedance response preserving the convolution form

The study of the thermal behavior of complex packages as multichip modules (MCM¿s) is usually carried out by measuring the so-called thermal impedance response, that is: the transient temperature after a power step. From the analysis of this signal, the thermal frequency response can be estimated, and consequently, compact thermal models may be extracted. We present a method to obtain an estimate of the time constant distribution underlying the observed transient. The method is based on an iterative deconvolution that produces an approximation to the time constant spectrum while preserving a convenient convolution form. This method is applied to the obtained thermal response of a microstructure as analyzed by finite element method as well as to the measured thermal response of a transistor array integrated circuit (IC) in a SMD package.

Analysis of Roman pottery graffiti by high resolution capture and 3D laser profilometry

The graffiti on pottery discovered on the site of Aventicum (Avenches, VD/Switzerland) form the largest corpus of minor inscriptions of the Roman Empire studied until now. Indeed, a total of 1828 graffiti have been found. The reading and the recording of the inscriptions are generally dependent on the state of conservation of the graffito and its support. In numerous cases, only a pale shadow of the inscription is visible, which makes traditional observations, such as visual observations with the naked eye, unsuitable for its decipherment. Consequently, advanced techniques have been applied for enhancing the readability of such inscriptions. In our paper we show the efficiency of 3D laser profilometry as well as high resolution photography as powerful means to decipher illegible engraved inscriptions. The use of such analyses to decipher graffiti on pottery or on other materials enables a better understanding of minor inscriptions and improves the knowledge of the daily life of ancient populations substantially.

How to assign a likelihood ratio in a footwear mark case: an analysis and discussion in the light of R v T

This article analyses and discusses issues that pertain to the choice of relevant databases for assigning values to the components of evaluative likelihood ratio procedures at source level. Although several formal likelihood ratio developments currently exist, both case practitioners and recipients of expert information (such as judiciary) may be reluctant to consider them as a framework for evaluating scientific evidence in context. The recent ruling R v T and ensuing discussions in many forums provide illustrative examples for this. In particular, it is often felt that likelihood ratio-based reasoning amounts to an application that requires extensive quantitative information along with means for dealing with technicalities related to the algebraic formulation of these approaches. With regard to this objection, this article proposes two distinct discussions. In a first part, it is argued that, from a methodological point of view, there are additional levels of qualitative evaluation that are worth considering prior to focusing on particular numerical probability assignments. Analyses will be proposed that intend to show that, under certain assumptions, relative numerical values, as opposed to absolute values, may be sufficient to characterize a likelihood ratio for practical and pragmatic purposes. The feasibility of such qualitative considerations points out that the availability of hard numerical data is not a necessary requirement for implementing a likelihood ratio approach in practice. It is further argued that, even if numerical evaluations can be made, qualitative considerations may be valuable because they can further the understanding of the logical underpinnings of an assessment. In a second part, the article will draw a parallel to R v T by concentrating on a practical footwear mark case received at the authors' institute. This case will serve the purpose of exemplifying the possible usage of data from various sources in casework and help to discuss the difficulty associated with reconciling the depth of theoretical likelihood ratio developments and limitations in the degree to which these developments can actually be applied in practice.

Analysis, modelling and classification of geospatial data using machine learning

The research considers the problem of spatial data classification using machine learning algorithms: probabilistic neural networks (PNN) and support vector machines (SVM). As a benchmark model simple k-nearest neighbor algorithm is considered. PNN is a neural network reformulation of well known nonparametric principles of probability density modeling using kernel density estimator and Bayesian optimal or maximum a posteriori decision rules. PNN is well suited to problems where not only predictions but also quantification of accuracy and integration of prior information are necessary. An important property of PNN is that they can be easily used in decision support systems dealing with problems of automatic classification. Support vector machine is an implementation of the principles of statistical learning theory for the classification tasks. Recently they were successfully applied for different environmental topics: classification of soil types and hydro-geological units, optimization of monitoring networks, susceptibility mapping of natural hazards. In the present paper both simulated and real data case studies (low and high dimensional) are considered. The main attention is paid to the detection and learning of spatial patterns by the algorithms applied.

BIO-INSPIRED COMPUTATIONAL TECHNIQUES APPLIED TO THE CLUSTERING AND VISUALIZATION OF SPATIO-TEMPORAL GEOSPATIAL DATA

The coverage and volume of geo-referenced datasets are extensive and incessantly¦growing. The systematic capture of geo-referenced information generates large volumes¦of spatio-temporal data to be analyzed. Clustering and visualization play a key¦role in the exploratory data analysis and the extraction of knowledge embedded in¦these data. However, new challenges in visualization and clustering are posed when¦dealing with the special characteristics of this data. For instance, its complex structures,¦large quantity of samples, variables involved in a temporal context, high dimensionality¦and large variability in cluster shapes.¦The central aim of my thesis is to propose new algorithms and methodologies for¦clustering and visualization, in order to assist the knowledge extraction from spatiotemporal¦geo-referenced data, thus improving making decision processes.¦I present two original algorithms, one for clustering: the Fuzzy Growing Hierarchical¦Self-Organizing Networks (FGHSON), and the second for exploratory visual data analysis:¦the Tree-structured Self-organizing Maps Component Planes. In addition, I present¦methodologies that combined with FGHSON and the Tree-structured SOM Component¦Planes allow the integration of space and time seamlessly and simultaneously in¦order to extract knowledge embedded in a temporal context.¦The originality of the FGHSON lies in its capability to reflect the underlying structure¦of a dataset in a hierarchical fuzzy way. A hierarchical fuzzy representation of¦clusters is crucial when data include complex structures with large variability of cluster¦shapes, variances, densities and number of clusters. The most important characteristics¦of the FGHSON include: (1) It does not require an a-priori setup of the number¦of clusters. (2) The algorithm executes several self-organizing processes in parallel.¦Hence, when dealing with large datasets the processes can be distributed reducing the¦computational cost. (3) Only three parameters are necessary to set up the algorithm.¦In the case of the Tree-structured SOM Component Planes, the novelty of this algorithm¦lies in its ability to create a structure that allows the visual exploratory data analysis¦of large high-dimensional datasets. This algorithm creates a hierarchical structure¦of Self-Organizing Map Component Planes, arranging similar variables' projections in¦the same branches of the tree. Hence, similarities on variables' behavior can be easily¦detected (e.g. local correlations, maximal and minimal values and outliers).¦Both FGHSON and the Tree-structured SOM Component Planes were applied in¦several agroecological problems proving to be very efficient in the exploratory analysis¦and clustering of spatio-temporal datasets.¦In this thesis I also tested three soft competitive learning algorithms. Two of them¦well-known non supervised soft competitive algorithms, namely the Self-Organizing¦Maps (SOMs) and the Growing Hierarchical Self-Organizing Maps (GHSOMs); and the¦third was our original contribution, the FGHSON. Although the algorithms presented¦here have been used in several areas, to my knowledge there is not any work applying¦and comparing the performance of those techniques when dealing with spatiotemporal¦geospatial data, as it is presented in this thesis.¦I propose original methodologies to explore spatio-temporal geo-referenced datasets¦through time. Our approach uses time windows to capture temporal similarities and¦variations by using the FGHSON clustering algorithm. The developed methodologies¦are used in two case studies. In the first, the objective was to find similar agroecozones¦through time and in the second one it was to find similar environmental patterns¦shifted in time.¦Several results presented in this thesis have led to new contributions to agroecological¦knowledge, for instance, in sugar cane, and blackberry production.¦Finally, in the framework of this thesis we developed several software tools: (1)¦a Matlab toolbox that implements the FGHSON algorithm, and (2) a program called¦BIS (Bio-inspired Identification of Similar agroecozones) an interactive graphical user¦interface tool which integrates the FGHSON algorithm with Google Earth in order to¦show zones with similar agroecological characteristics.