La régénération naturelle des peuplements de chênes verts (Quercus ilex) des systèmes agroforestiers de « dehesas » pose problème dans leur aire de distribution principale. La cause invoquée est l’absence de strate arbustive protégeant les semis de l’abroutissement et de la chaleur. La thèse, à travers la méthode régressive, propose d’explorer en détail cette piste. Le Parc Naturel de la Sierra de Grazalema (Andalousie, Espagne), en périphérie de la distribution des dehesas et inexploré sur ces questions, a été le lieu de nos investigations. La régénération naturelle a été analysée comme un concept évoluant dans le temps et les champs disciplinaires : c’est une technique naturalisée. A l’analyse de documents historiques, allant de 1765 à nos jours, s’ajoute un travail de terrain comprenant 71 transects mesurant les strates arbustive et arborée sur cinq terres publiques et trois terres privées. Une répartition en classes de diamètre pour la strate arborée a été effectuée et des variogrammes ont été utilisés pour analyser le patron de semis à plusieurs échelles. La distribution des classes de diamètre, suivant une distribution en « J inversé », indique une réussite de la régénération avec cependant une forte hétérogénéité entre les transect et le statut foncier. Un nombre important de jeunes arbres a été trouvés : 4516,65 semis/ha, 986 fourrés/ha et 543 gaules/ha. Les variogrammes indiquent des patrons de régénération à plusieurs échelles, une dizaine de mètres, 45 à 90 mètres, 100 à 200 mètres et proche du kilomètre. Ces résultats, très différents des études précédentes et venant compléter des trajectoires historiques, indiquent qu’il n’y a pas une forme de régénération mais plusieurs en fonction des époques mais aussi des milieux. Il y a des cas de régénération sans un passage par une « matorralisation », des cas s’inscrivant dans des dynamiques plus proches de séries (régressives ou progressives) et enfin des cas de régénération sensu stricto. / The regeneration of Holm oak (Quercus ilex) populations remains problematic in most dehesa agroforestry systems, allegedly due to the absence of a shrub layer protecting seedlings from grazing and heat. This thesis explores the latter claim using the regressive method. Our study focuses on the Sierra de Grazalema Natural Park, a site located at the periphery of dehesa regions in Andalusia, Spain and not previously analyzed in this manner. Regeneration was interpreted as a naturalized technique that evolves over time and within disciplinary fields. Our study features an analysis of historical documents, from 1765 to the present day, as well as fieldwork including 71 transects measuring shrub and tree layers on five public lands and three private lands. The tree layer was examined using a distribution into diameter classes and variograms analyzed seeding patterns on several scales. The reverse J-shaped diameter distribution indicates successful regeneration, but with a strong heterogeneity both among transects and due to land status. A significant number of young trees were found: 4516.65 seedlings / ha, 986 thickets / ha and 543 saplings / ha. Variograms indicate regeneration patterns present at several scales; at about ten meters, at 45 to 90 meters, at 100 to 200 meters and near one kilometer. These results, very different from previous studies, complement historical trajectories, and indicate that there is not one form of regeneration but several, related to both historical eras and environments. Patterns of regeneration include cases without a matorral shrublands and woodlands stage, cases belonging to a regressive or progressive series, and finally, cases of regeneration sensu stricto.

In this thesis we study the problem of estimation of parametric covariance and variogram functions for spatial and spatio- temporal random processes. It has the following principal parts. Variogram Estimation: We consider the "weighted" least squares criterion of fitting a parametric variogram function to second order stationary geo-statistical processes. Two new weight functions are investigated as alternative to the commonly used weight function proposed by Cressie (1985). We discuss asymptotic convergence properties of the sample variogram estimator and estimators of unknown parameters of parametric variogram functions, under a "mixed increasing domain" sampling design as proposed by Lahiriet al. While empirical results of Mean Square Errors, for parameter estimation, obtained using both the proposed functions are found to be comparatively better, we also theoretically establish that under general conditions one of the proposed weight functions give estimates with better asymptotic effciency. Spatio-Temporal Covariance Estimation: Over the past decade, there have been some important advances in methods for constructing valid spatiotemporal covariance functions; but not much attention has been given - so far - on methods of parameter estimation. In this thesis we propose a new frequency domain approach to estimating parameters of spatio-temporal covariance functions. We derive asymptotic strong consistency properties of the estimators using the concept of stochastic equicontinuity. The theory is illustrated with a simulation. Non-Linearity of Geostatistical Data: Linear prediction theory for spatial data is well established and substantial literature is available on the subject. Relatively less is known about non-linearity. In our final and ongoing, research problem we propose a non-linear predictor for geostatistical data. We demonstrate that the predictor is a function of higher order moments. This leads us to construct spatial bispectra for parametric third order moments.

No description available.

No description available.

Evapotranspiration (ET) is a key parameter for irrigation planning and management, and it is a crucial factor for water conservation practices considering the challenges associated with agricultural water availability. Field ET determination is the most accurate, but remains to be expensive and limited in scope. On the other hand, remote sensing is becoming an alternative tool for the estimation of ET. Operational ET algorithms, like the Moderate Resolution Imaging Spectroradiometer (MODIS)-based ET, are now successful at generating ET estimates globally at 1km resolution, however their intent is not management of agriculture irrigation. This research was done to develop an integrated method for downscaling MODIS ET appropriate for farm-level applications using geostatistical and remote sensing techniques. The proposed methodology was applied in the Wellton-Mohawk Irrigation and Drainage District of Yuma, Arizona. In a first effort, ET data was downscaled from standard 1-km-MODIS to a medium 250-m-spatial resolution via cokriging using Land Surface Temperature and Enhanced Vegetation Index as covariates. Results showed consistent downscaled ET with a variance greater than the variance of the coarse scale input and nearly similar mean values. This 250m product can serve larger irrigation districts in developed countries, where plot size is fairly large and regular. However, the size and shapes of most farms in developing countries makes the 250m ET challenging. For this reason, the second part of this work was done to downscale global scale 1km ET to 30m farm level application for irrigation use. This approach involved the generation of daily vegetation indices (VI) at 30m in order to support the downscaling of MODIS 1km ET. Landsat and MODIS reflectances were combined with the Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM) algorithm and the resulting VI data was used as a covariate to downscale ET with the cokriging approach. The results showed that the MODIS ET data seriously underestimates ET over irrigated areas. To correct this problem the MODIS data was then adjusted using field measured values to make it useful for operational purposes. The proposed geospatial method was applied to different growth stages of cotton and results were validated with actual ET from The Arizona Meteorological Network (AZMET) and published consumptive use of water for the area. The adjusted downscaled ET was comparable to these two published data (maximum error of 33%). This methodology is a practical alternative in areas where there is no ancillary data to estimate ET and it is expected to help in the planning of irrigation agriculture that will lead to improved agricultural productivity and irrigation efficiency.

No description available.

No description available.

Doctor of Philosophy / Department of Statistics / Juan Du / It is common to assume the spatial or spatio-temporal data are realizations of underlying random elds or stochastic processes. E ective approaches to modelling of the underlying autocorrelation structure of the same random eld and the association among multiple processes are of great demand in many areas including atmospheric sciences, meteorology and agriculture. To this end, this dissertation studies methods and application of the spatial modeling of large-scale dependence structure and spatio-temporal regression modelling. First, variogram and variogram matrix functions play important roles in modeling dependence structure among processes at di erent locations in spatial statistics. With more and more data collected on a global scale in environmental science, geophysics, and related elds, we focus on the characterizations of the variogram models on spheres of all dimensions for both stationary and intrinsic stationary, univariate and multivariate random elds. Some e cient approaches are proposed to construct a variety of variograms including simple polynomial structures. In particular, the series representation and spherical behavior of intrinsic stationary random elds are explored in both theoretical and simulation study. The applications of the proposed model and related theoretical results are demonstrated using simulation and real data analysis. Second, knowledge of the influential factors on the number of days suitable for fieldwork (DSFW) has important implications on timing of agricultural eld operations, machinery decision, and risk management. To assess how some global climate phenomena such as El Nino Southern Oscillation (ENSO) a ects DSFW and capture their complex associations in space and time, we propose various spatio-temporal dynamic models under hierarchical Bayesian framework. The Integrated Nested Laplace Approximation (INLA) is used and adapted to reduce the computational burden experienced when a large number of geo-locations and time points is considered in the data set. A comparison study between dynamics models with INLA viewing spatial domain as discrete and continuous is conducted and their pros and cons are evaluated based on multiple criteria. Finally a model with time- varying coefficients is shown to reflect the dynamic nature of the impact and lagged effect of ENSO on DSFW in US with spatio-temporal correlations accounted.

The demand for large scale storage of carbon dioxide (CO2) grows stronger as incentives to reduce greenhouse gas emissions are introduced. Geological storage sites such as depleted oil and gas reservoirs, unminable coal seams and deep saline water-saturated aquifers are a few of many possible geological storage sites. Geological formations offer large scale storage potential, hidden locations and are naturally occurring world wide. A disadvantage is the difficulty to investigate the properties of storage material over large areas. Reservoir simulation studies addressing issues of heterogeneous reservoirs are growing in number. There is still much to investigate however this study adds to the field by investigating the significance of the heterogeneity in hydraulic conductivity based on core sample data. The data was received from the main CO2 injection site Heletz, Israel in the European Union Seventh Framework Programme for research and technological development (EU FP7) project MUSTANG (CO2MUSTANG, 2011-03-13). By developing models using iTOUGH2/ECO2N, the aim of this study is to contribute to a better understanding of how the average permeability, variance in permeability and spatial correlation of the reservoir properties affect the distribution of CO2 within the deep saline aquifer target layer. In this study a stochastic simulation approach known as the Monte Carlo method is applied. Based on core sample data, geostatistical properties of the data are determined and utilized to create equally probable realizations where properties are described through a probability distribution described by a mean and variance as well as a constructed semivariogram. The results suggest that deep saline aquifers are less storage effective for higher values of average permeability, variance in permeability and spatial correlation. The results also indicate that the Heletz aquifer, with its highly heterogeneous characteristics, in some extreme cases can be just as storage effective as a deep saline aquifer ten times as permeable consisting of homogeneous sandstone. / Incitament för minskningar av växthusgaser har på senare tid ökat efterfrågan för storskalig lagring av koldioxid (CO2). Geologiska lagringsplatser som exploaterade olje- och gasreservoarer, svårutvunna kollager och djupt belägna salina akvifärer är exempel på potentiella lagringsplatser. Sådana geologiska formationer erbjuder storskalig lagring, dold förvaring och är naturligt förekommande världen över. Dock finns det stora svårigheter i att undersöka de materiella egenskaperna för hela lagringsområden. Simuleringsstudier som hantera frågor gällande reservoarers heterogenitet växer i antal. Det finns fortfarande mycket kvar att undersöka och denna studie bidrar till detta forskningsområde genom att undersöka betydelsen av heterogenitet i hydraulisk konduktivitet för spridningen av koldioxid med hjälp av uppmätt brunnsdata. Data erhölls från lagringsplatsen Heletz i Israel som är den huvudsakliga lagringplatsen i projektet MUSTANG är en del av den Europeiska Unionens sjunde ramprogram för forskning och teknisk utveckling (EU FP7), (CO2MUSTANG, 2011/3/13). Genom att utveckla modeller med hjälp av programvaran iTOUGH2/ECO2N är syftet med denna studie att bidra till en bättre förståelse för hur den genomsnittliga permeabilitet, varians i permeabilitet samt rumslig korrelation av reservoaregenskaper påverkar fördelningen av CO2 i den djupa saltvattenakvifären Heletz. Denna studie använde sig av stokastisk simulering genom att tillämpa Monte Carlo-metoden. Med hjälp av tidigare uppmätt brunnsdata kunde geostatistiska egenskaper bestämmas för att skapa ekvivalent sannolika realiseringar. De geostatistiska egenskaperna beskrevs med en sannolikhetsfördelning genom medelvärde och varians samt ett konstruerat semivariogram. Resultaten tyder på att djupa saltvattenakvifärer är mindre lagringseffektiva vid högre värden av genomsnittlig permeabilitet, varians i permeabilitet och rumslig horisontell korrelation. Resultaten visar även att Heletz akvifär, med dess mycket heterogena egenskaper, i extrema fall kan vara lika lagringsineffektiv som en djupt belägen saltvattenakvifär med tio gånger högre genomsnittlig permeabilitet.

Made available in DSpace on 2016-08-17T14:53:15Z (GMT). No. of bitstreams: 1 Daniel Rodrigues Ericeira.pdf: 2002346 bytes, checksum: df76ac081a5d0e5816a81b5699935561 (MD5) Previous issue date: 2011-03-17 / Mammography is the exam of the breast, used as breast cancer prevention and also as a diagnostic method. This exam, which consists in an X-Ray of the breast, allows cancer detection. The purpose of this work is to use image processing techniques and computer vision to help specialists in detecting suspect regions and masses in digital mammographies. The first stage of the methodology consists in pre-processing the images to make them more suitable to registration, through noise reduction, image segmentation and re-scale. The next stage presents bilateral left and right breast image pairs registration. In order to correct position and compression differences that occur during the exams, rigid registration (followed by optic flow deformable registration) was applied in each image pair. Corresponding pairs of regions were related and their mutual variations were measured through cross-variogram spatial description. On the next stage, a training model for a Support Vector Machine (SVM) was created using as characteristics the cross-variogram values of each pair of regions of 180 cases. This SVM was tested for 100 new cases. The region pairs that contained lesions were classified as suspect regions , and the other regions as non-suspect regions . From the suspect regions, variogram characteristics were extracted as tissue texture descriptors. The regions that contained masses were classified as mass regions and the other regions as non-mass regions . Stepwise linear discriminant analysis was applied to select the most significant characteristics to train the second SVM. Tests with 30 new cases were performed for the trained SVM final classification in mass or non-mass . The best case presented on the final classification: 96% accuracy, 100% sensitivity and 95,34% specificity. The worst case presented: 70% accuracy, 100% sensitivity and 67,56% specificity. On average, the 30 cases presented: 90% accuracy, 100% sensitivity and 85% specificity. / A mamografia é um exame de mama, utilizado de forma preventiva ao câncer de mama e também como método diagnóstico. Este exame, que consiste em uma radiografia das mamas, permite a detecção do câncer. O objetivo deste trabalho é utilizar técnicas de processamento de imagens e visão computacional para auxiliar especialistas na detecção de regiões suspeitas e detecção de massas mamárias em mamografias digitais. A primeira etapa da metodologia consiste em pré-processar as imagens de forma a torná-las mais apropriadas ao registro, através de redução de ruído, segmentação e re-dimensionamento. A etapa seguinte apresenta o registro bilateral de pares de mamas esquerda e direita. Para corrigir as diferenças de posicionamento e compressão ocorridas no momento do exame, o método de registro rígido foi aplicado (seguido do método de registro deformável com fluxo óptico) para cada par de imagens. Pares de regiões correspondentes foram relacionados e suas variações foram medidas através do descritor espacial variograma cruzado. Na etapa seguinte, foi criado um modelo para treinamento de uma Máquina de Vetores de Suporte (MVS) utilizando como características os valores de variograma cruzado de cada par de janelas de 180 casos. Esta MVS foi testada em 100 novos casos. Os pares que continham lesões foram classificados como regiões suspeitas ; as demais, como regiões não-suspeitas . Destas regiões suspeitas, foram extraídas características de variograma como descritores de textura de tecido. As regiões que continham massas foram classificadas como regiões de massa e as demais como regiões de não-massa . Análise linear discriminante stepwise foi aplicada para selecionar as características mais significativas para treinamento de uma segunda MVS. Foram realizados testes com 30 novos casos para a classificação final pela MVS treinada em massa e nãomassa . O melhor resultado apresentou na classificação final: 96% de acurácia, 100% de sensibilidade e 95,34% de especificidade. O pior caso apresentou: 70% de acurácia, 100% de sensibilidade e 67,56% de especificidade. Em média, os 30 casos apresentaram: 90% de acurácia, 100% de sensibilidade e 85% de especificidade.