Дисертації з теми "Neige – Télédétection"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-35 дисертацій для дослідження на тему "Neige – Télédétection".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Bourdelles, Barbara. "Etude des caractéristiques de surface de la neige par télédétection visible et infrarouge." Phd thesis, Université Joseph Fourier (Grenoble), 1994. http://tel.archives-ouvertes.fr/tel-00759878.
Повний текст джерелаHaddjeri, Ange. "Modélisation de l'évolution de la neige soufflée et évaluation de la variabilité spatiale induite." Electronic Thesis or Diss., Université de Toulouse (2023-....), 2024. http://www.theses.fr/2024TLSES085.
Повний текст джерелаAlpine snow cover is highly variable both spatially and temporally. An accurate knowledge of this variability is a high stake for water supply, hydropower production, and hazard forecasting such as avalanches and floods. Snowpack observation and numerical simulation are complementary tools for these applications. In France, the ISBA-Crocus snow model is operated on a daily basis but currently only provides a large-scale assessment of the snow conditions. The future evolution of this system will rely on a 250 m horizontal resolution to better describe the spatial variability of the snow cover. This resolution requires the representation of additional processes such as lateral wind redistribution for realistic simulations at high elevations. As with similar snow modeling systems in other countries, the evaluation of these spatialized regional simulations is still challenging due to the sparsity of observations and various interacting uncertainties and processes contributing to spatial variability at this scale. In this context, the objective of this PhD is to develop and evaluate a spatialized alpine snow simulation system with wind lateral redistribution. Emphasis will be placed on the evaluation methods required to evaluate snow regional simulations with satellite observations, which should also benefit similar snow simulation systems in mountain environments. In the first part, we present the design and development of the SnowPappus blowing snow model, coupled with the ISBA-Crocus simulation system over 2-dimensional simulation domains with targeted applications covering large spatial and temporal extents (all French mountain ranges and several years). The SnowPappus model simulates blowing snow occurrence, horizontal transport flux, and sublimation rate as a function of 2D atmospheric forcing and snow surface parameters. Erosion and accumulation are then obtained from an upwind scheme of mass balance. Point-scale evaluations of snow occurrence detection and blowing snow fluxes showed that SnowPappus performs as well as the larger-scale SYTRON scheme while adding access to spatialized information. Then, we evaluate spatialized simulations of the SnowPappus model over a 902 km² region in the French Alps with satellite images during three snow seasons. We compared snow cover simulations to the spatial distribution of snow height obtained from Pleiades satellites stereo-imagery and to Snow Melt-Out Dates derived from Sentinel-2 optical images. The sensitivity of simulations to three different precipitation datasets and two horizontal resolutions is also analyzed. Our results show that the SnowPappus model enhances the snow cover spatial variability at high elevations and near peaks and ridges. Our study shows the necessity to consider error contributions from precipitation forcing and the unresolved subgrid variability for robust evaluations of spatialized snow simulations. Finally, we tested and adapted Fuzzy verification methods, developed for atmospheric simulation evaluations to snow use cases. This kind of verification method is more challenging than spatial distribution analysis and helps to assess the spatial agreements between observations and simulations. Although fuzzy verification techniques can help to better qualify the spatial agreement between simulations and observations, they have limitations over complex topography and do not allow for disentangling error from true localization errors with intensity error compensation. Which is a strong limitation for real use cases of alpine snow forecast verification. The methodological advances of this work help to identify the strengths and weaknesses of high-resolution snow simulations, including the added value of an explicit representation of blowing snow. This is an important step in guiding the use of such simulations in all target applications among winter mountain stakeholders
Masson, Théo. "Fusion de données de télédétection haute résolution pour le suivi de la neige." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAT112/document.
Повний текст джерелаRemote sensing acquisitions have complementary characteristics in terms of spatial and temporal resolution and can measure different aspects of snow cover (e.g., surface physical properties and snow type). By combining several acquisitions, it should be possible to obtain a precise and continuous monitoring of the snow. However, this task has to face the complexity of processing satellite images and the possible confusion between different materials observed. In particular, the estimation of fractional information, i.e., the amount of snow in each pixel, requires to know the proportion of the materials present in a scene. These proportions can be obtained performing spectral unmixing. The challenge is then to effectively exploit the information of different natures that are provided by the multiple acquisitions in order to produce accurate snow maps.Three main objectives are addressed by this thesis and can be summarized by the three following questions:- What are the current limitations of state-of-the-art techniques for the estimation of snow cover extent from optical observations?- How to exploit a time series for coping with the spectral variability of materials?- How can we take advantage of multimodal acquisitions from optical sensors for estimating snow cover maps?A complete study of the various snow products from the MODIS satellite is proposed. It allows the identification of numerous limitations, the main one being the high rate of errors during the estimation of the snow fraction (approximately 30%).The experimental analysis allowed to highlight the sensitivity of the spectral unmixing methods against the spectral variability of materials.Given these limitations, we have exploited the MODIS time series to propose a new endmembers estimation approach, addressing a critical step in spectral unmixing. The low temporal evolution of the medium (except snow) is then used to constrain the estimation of the endmembers not only on the image of interest, but also on images of the previous days. The effectiveness of this approach, although demonstrated here, remains limited by the spatial resolution of the sensor.Data fusion has been considered aiming at taking advantage of multiple acquisitions with different characteristics in term of resolution available on the same scene. Given the limitations of the actual methods in the case of multispectral sensors, a new fusion approach has been proposed. Through the formulation of a new model and its resolution, the fusion between optical sensors of all types can be achieved without consideration of their characteristics. The various experiments on the estimation of snow maps show a clear interest of a better spatial resolution to isolate the snow covered areas. The improvement in spectral resolution will improve future approaches based on spectral unmixing.This work explores the new possibilities of development for the observation of snow, but also for the combined use of the satellite images for the observation of the Earth in general
Köhn, Jacqueline. "Caractérisation de la température de la neige par télédétection micro-onde passive au Canada." Mémoire, Université de Sherbrooke, 2006. http://savoirs.usherbrooke.ca/handle/11143/2502.
Повний текст джерелаMazeh, Fatme. "Modélisation Numérique de la Réponse du Radar à la Neige pour Mesurer sa Profondeur avec la Technique de la Rétrodiffusion Multiple." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAT055/document.
Повний текст джерелаStudy of snow is an important domain of research in hydrology and meteorology. It has been demonstrated that snow physical properties can be retrieved using active microwave sensors. This requires an understanding of the interaction between electromagnetic (EM) waves with natural media. The objective of this work is two-fold: to study numerically all physical forward models concerning the EM wave interaction with snow and to develop an inverse scattering algorithm to estimate snow depth based on radar backscattering measurements at different frequencies and incidence angles. For the first part, the goal is to solve the scattering calculations by means of the well-known electromagnetic simulator Ansoft High Frequency Structure Simulator (HFSS). The numerical simulations include: the effective permittivity of snow, surface scattering phenomena in layered homogeneous media (air-snow-ground) with rough interfaces, and volume scattering phenomena when treating snow as a dense random media. So, the critical issue for the first part of this thesis is testing the validity of theoretical models through a careful numerical setup.For the second part, the study is extended to develop a retrieval method to estimate snow thickness over ground from backscattering observations at L- and X-band (1.5 and 10 GHz) using multiple incidence angles. The return signal from snow over ground is influenced by: surface scattering, volume scattering, and the noise effects of the radar system. So, the backscattering coefficient from the medium is modelled statistically by including a white Gaussian noise (WGN) into the simulation. This inversion algorithm involves two steps. The first is to estimate snow density using L-band co-polarized backscattering coefficient at normal incidence. The second is to estimate snow depth from X-band co-polarized backscattering coefficients using two different incidence angles. For a 0.02 noise variance, all retrieved values have an error less than 2% for a snow depth range of [50-300] cm. This algorithm was verified by simulating using Agilent’s SystemVue electronic system level design software
Leroux, Catherine. "Etude théorique et expérimentale de la réflectance de la neige sur le spectre solaire : application à la télédétection." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 1996. http://tel.archives-ouvertes.fr/tel-00781634.
Повний текст джерелаGalligani, Victoria Sol. "La radiométrie micro-onde et millimétrique pour la caractérisation et quantification des nuages de glace et de la neige." Observatoire de Paris, 2014. https://hal.science/tel-02095287.
Повний текст джерелаQuantification of the cloud frozen phase on a global basis is essential to fully capture and quantify the Earth energy budget and hydrological cycle. The estimation of frozen quantities from satellite remote sensing, however, is at a very early stage. The main reason is the complex variability of the cloud frozen phase and the lack of parameterizations of their microphysical properties, and thus radiative properties. This thesis contributes to the development of the ice cloud remote sensing, by providing a better understanding of the sensitivity of microwave and millimeter satellite observations to the microphysical properties of the frozen phase, specifically snow. Current microwave and millimeter observations are interpreted via radiative transfer simulations, mainly for passive observations, but active measurements are also considered. Two main studies are pursued: (1) the analysis and interpretation of specific polarized scattering signatures over ice and snow clouds, and (2) the simulation of realistic passive and active microwave responses over ice and snow clouds, and their evaluation with satellite observations. Polarized observations are carefully analyzed with ancillary data and are interpreted with radiative transfer simulations, including the first polarized passive observations above 100GHz with Megha-Tropiques. Finally, the radiative transfer model is coupled to a meso- scale cloud model to simulate consistently coincident active and passive observations of real scenes, and assess the sensitivity of active and passive simulations to the different microphysical parameters
Vachon, François. "Estimation de l'équivalent en eau de la neige en milieu subarctique du Québec par télédétection micro-ondes passives." Thèse, Université de Sherbrooke, 2009. http://savoirs.usherbrooke.ca/handle/11143/2806.
Повний текст джерелаChagnon, Frédéric. "Caractérisation des états de surface par télédétection infrarouge thermique multispectrale contribution à l'étude des conditions de viabilité hivernale." Thèse, Université de Sherbrooke, 2008. http://savoirs.usherbrooke.ca/handle/11143/2783.
Повний текст джерелаBusseau, Bruno-Charles. "Analyse des effets de la végétation sur le couvert de neige dans la zone de transition arctique-subarctique par mesures in-situ et télédétection optique (Nunavik)." Mémoire, Université de Sherbrooke, 2017. http://hdl.handle.net/11143/10160.
Повний текст джерелаAbstract : Recent studies have shown that northern vegetation has been growing in relation to a warming climate over the last four decades, especially across the transition zone between tundra and taiga. Shrub growth affects snow properties and the surface energy budget, which must be better studied to quantify shrub-snow-climate feedbacks. The objective of this research is to improve the characterization of the impact of shrubs on snow evolution, from its accumulation to its melt, using in-situ and satellite measurements. The research is presented for the Umiujaq site, Nunavik, representative of the low Arctic – Subarctic transition zone. Snow depth, measured along numerous transects spanning different land cover types is found to increase by a factor 2.5 to 3 between tundra and forest, while snow density decreases. This illustrates the trapping effect of vegetation well. Complementary continuous snow depth measurements using weather stations from two sites (tundra with low shrubs and a small clearing with shrubs within the forest) show different site-dependent behaviors. Spatial analysis from high-resolution Pleiades images combined with Landsat (Normalized Difference Snow Index) and MODIS (Fractional Snow Cover) images suggest a slight delay in melt over open and dense forest areas compared to tundra and dense high shrubs. A technic to measure snow depth using high resolution is also discussed.
Surdyk, Sylviane. "Etudes des signatures spectrales micro-ondes obtenues par télédétection sur la calotte polaire antarctique : comparaison avec des données de terrain et modélisation de l'émissivité de la neige." Phd thesis, Grenoble INPG, 1993. http://tel.archives-ouvertes.fr/tel-00765202.
Повний текст джерелаRoy, Alexandre. "L'apport de la télédétection à un modèle de neige appliqué à un système d'aide à la gestion des barrages dans le sud du Québec." Mémoire, Université de Sherbrooke, 2009. http://savoirs.usherbrooke.ca/handle/11143/2629.
Повний текст джерелаKohn, Jacqueline. "Inversion des observations spatiales micro-ondes pour la détermination de la température du sol en présence de neige." Thèse, Université de Sherbrooke, 2009. http://savoirs.usherbrooke.ca/handle/11143/2812.
Повний текст джерелаPivot, Frédérique. "Télédétection appliquée au suivi des variations spatio-temporelles du couvert nival à la limite des arbres (Churchill-Manitoba)." Lille 1, 2000. https://pepite-depot.univ-lille.fr/RESTREINT/Th_Num/2000/50377-2000-25.pdf.
Повний текст джерелаPhan, Xuan Vu. "Assimilation de données radar satellitaires dans un modèle de métamorphisme de la neige." Thesis, Rennes 1, 2014. http://www.theses.fr/2014REN1S132/document.
Повний текст джерелаCharacterization of snowpack structure is an important issue for the management of water resources and the prediction of avalanche risks. New Synthetic Aperture Radar (SAR) satellites in X-band at high-resolution allow us to acquire image data with metric resolution and daily observations. In this work, an electromagnetic backscattering model applicable for dry snow is adapted for X-band and higher frequencies. The 3D-VAR data assimilation algorithm is then implemented to constrain the evolution of the snow metamorphisme model SURFEX/Crocus using satellite observations. Finally, the algorithm is evaluated using image data acquired from TerraSAR-X satellite on the Argentiere glacier in the Chamonix Valley of the French Alps. This first comparison shows the high potential of the data assimilation assimilation method using X-band SAR data for characterization of the snowpack
Bergeron, Jean. "L’assimilation de données multivariées par filtre de Kalman d’ensemble pour la prévision hydrologique." Thèse, Université de Sherbrooke, 2017. http://hdl.handle.net/11143/10540.
Повний текст джерелаQuense, Jorge. "Changement climatique et dynamique de la végétation dans les Andes du Chili central, depuis le milieu du XXème siècle : l'exemple de la vallée de Yerba Loca." Phd thesis, Université de Grenoble, 2011. http://tel.archives-ouvertes.fr/tel-00639115.
Повний текст джерелаDurán, Alárcon Claudio. "Ground-based remote sensing of Antarctic and Alpine solid precipitation." Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAU024/document.
Повний текст джерелаSolid precipitation plays an important role in the Earth's climate system, as well as for the maintenance of ecosystems and the development of human society. The large uncertainty in precipitation estimates and the discrepancies within climate model projections make this component of the hydrological cycle important as a research topic. Remote sensing allows to monitor precipitation and clouds in regions where in-situ observations are scarce and scattered, but with limited temporal resolution and a blind zone close to the ground level for spaceborne sensors, and limited visibility in the lower atmosphere in complex terrain for ground-based radars. The objectives of this dissertation are the following: 1) to characterize cloud and precipitation in Antarctica, detecting the presence of supercooled liquid and ice particles near the ground level using a ground-based 532-nm depolarization lidar; 2) to characterize the vertical structure of the precipitation in two contrasted but important regions of the cryosphere, Antarctica and the Alps, in the low troposphere using ground-based radars.In this study, a cloud and precipitation hydrometeor detection method is proposed using lidar data, complemented with a K-band micro rain radar (MRR) to improve the detection of precipitation, both instruments deployed at the Dumont d'Urville (DDU) station in East Antarctica. A method based on lidar depolarization and attenuated backscattering coefficient and the use of k-means clustering is developed for the particle classification. The classification of cloud and precipitation particles provides the vertical distribution of supercooled liquid water, as well as planar oriented ice and randomly oriented ice particles. The comparison between ground-based and satellite-derived classifications shows consistent patterns for the vertical distribution of supercooled liquid water in clouds.The vertical structure of precipitation near the surface is analyzed using the Doppler moments derived from three MRR profiles at DDU, the Princess Elisabeth (PE) station, at the interior of East Antarctica, and at the Col de Porte (CDP) station, in the French Alps. These analyses demonstrate that local climate plays an important role in the vertical structure of the precipitation. In Antarctica, the strong katabatic winds blowing from the high plateau down to the coast decrease the radar reflectivity factor near the surface due to the sublimation of the snowfall particles. Doppler moments also provide rich information to understand precipitation processes, such as aggregation and riming, as observed at DDU and CDP.The results also show that in the interior of the Antarctic continent a significant part (47%) of the precipitation profiles completely sublimate before reaching the surface, due to the dry atmospheric conditions, while in the coast of Antarctica it corresponds to about the third part (36%). In the Alps, this percentage is reduced to 15%. The major occurrence of particle sublimation is observed below the altitude where CloudSat profiles are contaminated by ground clutter. Therefore, this phenomenon cannot be fully captured from space with the current generation of sensors.This dissertation contributes to the study of the vertical structure of snowfall in the low troposphere, useful for the evaluation of precipitation remote sensing products, which may have severe limitations in the vicinity of the surface
Dupont, Florent. "Télédétection micro-onde de surfaces enneigées en milieu arctique : étude des processus de surface de la calotte glaciaire Barnes, Nunavut, Canada." Phd thesis, Université de Grenoble, 2013. http://tel.archives-ouvertes.fr/tel-01070037.
Повний текст джерелаDeschamps-Berger, César. "Apport de la photogrammétrie satellite pour la modélisation du manteau neigeux." Thesis, Toulouse 3, 2021. http://www.theses.fr/2021TOU30044.
Повний текст джерелаMountain snowpack is a major resource for ecosystems and human activities. It supplies water for crop irrigation, human consumption, hydropower industries and the tourism sector. It is also a cause of damage in avalanche prone areas. The monitoring and study of mountain snowpack usually rely on field measurement networks, close range remote sensing and modeling. Recent improvements in satellite photogrammetry provide an alternative to measure the high spatial variability of the snowpack, which cannot be sampled by automatic networks. The results presented here, contribute to improve the mapping of snow-depth in mountains with satellite photogrammetry, a key variable for hydrology and risk assessment. Snow-depth maps from pairs and triplets of stereo images of the Pléiades satellite are calculated at several sites. The comparison with a reference snow-depth map measured with airborne lidar in California (USA), provides a robust estimation of the satellite products error. At the 3 m pixel scale, the standard error is about 0.7 m. The error decreases to 0.3 m when the snow-depth maps are averaged over areas greater than 103 m2. With this accuracy, Pléiades snow-depth maps allow the observation of the processes modeling mountain snowpack (wind transport, avalanche), the measurement of the snow volume over a 100 km2 area and the description of the spatial variability of the snowpack. The assimilation of such satellite snow-depth maps in the SAFRAN-Crocus snowpack model, resulted in promising outcomes for a mountainous catchment in the Pyrenees. A particle filter is used on a regular grid with 250 m spacing over five winters with one assimilation date per winter, near peak accumulation. The assimilation corrects an underestimation of the precipitation in the meteorological forcings. It also introduces spatial variability otherwise lacking in the forcings and the processes modeled. This innovative use of satellite products and complex spatial modeling, could help address the challenge of estimating snow distribution in the world's mountains
Brucker, Ludovic. "Modélisation de l'émission micro-onde du manteau neigeux : applications en Antarctique et au Québec." Phd thesis, Université Joseph Fourier (Grenoble), 2009. http://tel.archives-ouvertes.fr/tel-00433824.
Повний текст джерелаLa télédétection passive, en particulier dans le domaine spectral des micro-ondes, est adaptée à l'interprétation et au suivi des propriétés physiques du manteau neigeux. Effectivement, le rayonnement micro-onde émane du sol ou du manteau neigeux lui-même, puis se propage jusqu'à la surface. Ainsi, le rayonnement émergeant contient de l'information sur les variations verticales des propriétés de la neige, telles que la température ou les propriétés de microstructure (taille de grains et densité). Ces trois propriétés de la neige déterminent l'émission micro-onde d'un manteau sec. Lorsqu'il est humide, la teneur en eau liquide devient par contre la propriété dominant l'émission. Les évolutions temporelles et les variations verticales de ces différentes propriétés sont définies par la métamorphose. Leur lien avec l'émission micro-onde est considéré dans le transfert radiatif.
Cette thèse a pour objectif d'expliquer l'émission micro-onde de la neige par voie de modélisation afin de comprendre l'évolution des principales propriétés physiques de la neige. Le transfert radiatif dans la neige a été calculé avec les modèles multicouches Microwave Emission Model of Layered Snowpacks (MEMLS) et MultiLayered Dense Media Radiative Transfer (DMRT-ML), s'appuyant sur des approches respectivement semi-empirique et théorique. Les profils stratigraphiques de la neige utilisés en entrée ont été mesurés, estimés de façon aléatoire, modélisés avec une relation simple de la métamorphose ou avec le modèle d'évolution thermodynamique de la neige Crocus.
Ces modèles et approches ont été appliqués sur deux types de manteau neigeux, permanent en Antarctique et saisonnier au Québec. Dans le premier cas, l'évolution temporelle de la température de brillance a été modélisée localement, à Dôme C, à partir de mesures in situ des propriétés de la neige. Dans cette approche, l'émissivité est modélisée à partir de mesures et reste par conséquent applicable localement. Pour modéliser l'émissivité à l'échelle de l'Antarctique, différents profils synthétiques de taille de grains et de densité ont été testés. Dans tous les cas, la variation verticale de la taille de grains est apparue déterminante pour prévoir l'émissivité en polarisation verticale. Cette sensibilité a été exploitée pour estimer à l'échelle du continent cette variable glaciologique importante. Le profil de densité et les propriétés de surface déterminent quant à eux l'écart entre les polarisations verticale et horizontale.
L'émission micro-onde d'un manteau saisonnier au Québec a également été abordée. La spécificité de l'étude est de prévoir l'évolution temporelle de la température de brillance avec un modèle d'évolution thermodynamique de la neige couplé à un modèle de transfert radiatif micro-onde, ici Crocus-MEMLS. Cette approche a permis d'interpréter finement l'évolution temporelle des températures de brillance mesurées avec un radiomètre au sol et de mettre en doute certaines relations physiques du modèle Crocus. Les résultats ont mis en évidence la complexité du signal micro-onde pour des manteaux évoluant rapidement à des températures proches du point de fusion.
Edel, Léo. "Vers une meilleure connaissance des précipitations en Arctique : utilisation de la télédétection spatiale micro-ondes." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLX108.
Повний текст джерелаBecause solid precipitation measurements at the surface are complex in Arctic, snowfall rates present significant differences between recent datasets. To further characterize arctic precipitation, microwave remote sensing is an appropriate tool. The radar observations onboard CloudSat provides rates of snowfall at the surface for a 4-years period. These retrievals are validated with in situ data, and compared to various datasets. Despite a good qualitative agreement, significant differences are observed, especially over Greenland. The regional reanalysis shows a better agreement with CloudSat retrievals than the global reanalysis, especially regarding the seasonnal distribution of snowfall rates. Then, CloudSat observations are used as a reference to evaluate the ability of passive microwave sounders to detect arctic snowfall for frequencies around 183 GHz. Detection is possible and relies mainly on brightness temperatures at 190 and 183 ± 3 GHz as well as the temperature near the surface and the integrated water vapor. A poor detection capability is observed in cold conditions and for light snowfall. Despite these limitations, the algorithm provides significant information for intense snowfalls, with good sampling due to its wide swath and long time series. Available for the last 20 years, passive microwave observations show a notable ability for a better characterization of arctic snowfall
Dumont, Marie. "Détermination de l'albédo des surfaces enneigées par télédétection : application à la reconstruction du bilan de masse du glacier de Saint Sorlin." Phd thesis, Université Paris-Est, 2010. http://tel.archives-ouvertes.fr/tel-00582925.
Повний текст джерелаRondeau-Genesse, Gabriel. "Suivi de l'eau liquide dans la neige par images radar en bande C et par modélisation fine du manteau neigeux." Mémoire, Université de Sherbrooke, 2015. http://hdl.handle.net/11143/8140.
Повний текст джерелаCluzet, Bertrand. "Application des approches d'ensemble à l'assimilation de données télédétectées dans des simulations numériques spatialisées du manteau neigeux saisonnier." Electronic Thesis or Diss., Toulouse 3, 2020. http://www.theses.fr/2020TOU30330.
Повний текст джерелаUnderstanding mountain snowpack variability is key to anticipate avalanche hazards and monitor water resources. On the one hand, remotely-sensed and in-situ observations of snow have a limited spatial and temporal coverage. For instance, visible and near infrared satellite reflectances provide useful information on snowpack surface properties, but are affected by important gaps of coverage e.g. due to clouds. Likewise, in-situ observations of the height of snow (HS) are reliable but with a limited representativeness and spatial coverage. On the other hand, detailed snowpack models can simulate the complete snow stratigraphy virtually anywhere, but they suffer from large uncertainties in their meteorological inputs and their representation of snow physical processes. Thus, data assimilation offers an unique opportunity to merge information from observations and models into a better estimate of the snowpack state. The aim of this thesis is to investigate the potential for satellite reflectances and in-situ HS to improve snowpack simulations in mountainous areas via assimilation. In this work, we will try to address the following questions: • Can we use observations of snowpack reflectance from satellites to better constrain snowpack modelling over mountainous areas? • Can we propagate information on the snowpack state from observed areas to unobserved areas with data assimilation? • To what extent can we use in-situ observations of HS to improve snowpack simulations in their neighborhood? We opt for a sequential ensemble data assimilation strategy, using the Particle Filter algorithm (PF), which is well adapted to detailed snowpack models. An ensemble modelling system is built by forcing ESCROC, a multiphysics ensemble of snowpack models, with an ensemble of stochastic perturbations on SAFRAN meteorological analyses. This design enables the modelling system to account for its main sources of uncertainty. Several innovative versions of the PF are developed in order to assimilate large numbers of observations and propagate information to unobserved areas while avoiding PF degeneracy, an issue arising when the number of observations increases. The potential for assimilation of satellite reflectance is assessed by comparing MODIS observations with simulated reflectances. We conduct twin experiments assimilating partial observations to analyse the ability of the PF to propagate information into unobserved areas. Finally, we assess the added value of the assimilation of HS observations from an observation network over the Alps and Pyrenees using a Leave-One-Out approach. Results show that the proposed methodology is efficient to tackle PF degeneracy while managing to propagate information across topographic conditions. Though standard MODIS observations cannot be directly assimilated because they are biased, the assimilation of HS observations have some added value where modelling errors are systematic and larger thaniv natural variability. This work is a novel contribution to improve the assimilation of other satellite products and in-situ HS observations in a spatialised context, a significant qualitative leap for avalanche forecasting and hydrological studies
Lacroix, Pascal. "Apport de l'altimétrie radar spatiale à l'étude de la neige de la calotte polaire Antarctique." Phd thesis, Université Paul Sabatier - Toulouse III, 2007. http://tel.archives-ouvertes.fr/tel-00216105.
Повний текст джерелаDepuis 2002 et le lancement de ENVISAT, on dispose d'un altimètre radar qui couvre 80 \% de la calotte polaire Antarctique, dont la particularité est d'acquérir des signaux à deux fréquences différentes (bande S à 3.2 GHz et bande Ku à 13.6 GHz). Ces deux ondes pénètrent dans le manteau neigeux sur plusieurs mètres et ont des sensibilités aux propriétés de la neige différentes. Ainsi, l'idée de cette thèse est d'utiliser cette double information pour retrouver les propriétés du manteau neigeux.
On se propose de résoudre cette problématique par une analyse et une modélisation des signaux altimétriques bi-fréquences sur la calotte polaire, puis par leur inversion. On se penche tout d'abord sur quelques études de cas pour estimer la sensibilité des signaux aux différentes propriétés de la neige: i/ On montre tout d'abord que le signal altimétrique est sensible à la rugosité de la surface à différentes échelles, puis ii/ que le signal altimétrique est sujet à des variations saisonnières causées par la densification de la neige en surface, et enfin iii/ que les ondes radars sont réfléchies par des strates en profondeur.
Un modèle de l'interaction de l'onde avec le manteau neigeux est réalisé simultanément aux deux fréquences, afin de permettre une comparaison de ces signaux entre eux. Les résultats du modèle sont utilisés pour expliquer les variations saisonnières précédemment observées. Finalement, les paramètres du manteau neigeux sont estimés à l'échelle de la calotte polaire antarctique. Les tailles de grains retrouvées présentent un grossissement vers l'intérieur du continent. La densité montre des variations saisonnières de plusieurs g.cm3 notamment sur les côtes antarctiques. Certaines régions présentent un état de surface de la neige particulièrement lisse (Dronning Maud Land, par exemple).
La donnée in situ de l'état de surface de la neige étant quasi inexistante sur les calottes polaires, on développe finalement un protocole de mesure de la rugosité de la neige, qui est testé sur un glacier du Spitzberg.
Thériault, Nathalie. "Analyse de sensibilité et amélioration des simulations d’albédo de surfaces enneigées dans les zones subarctiques et continentales humides à l’est du Canada avec le schéma de surface CLASS." Mémoire, Université de Sherbrooke, 2015. http://hdl.handle.net/11143/6946.
Повний текст джерелаAbstract : The surface energy balance of northern regions is closely linked to surface albedo (fraction of solar radiation reflected by a surface) variations. These variations are strongly influenced by the presence, depth and physical properties of the snowpack. Climate change affects significantly snow cover evolution, and decreases surface albedo and snow albedo with positive feedback to climate. Despite the importance of the albedo, many models empirically compute it, which can induce significant biases with albedo observations depending on studied surfaces. The Canadian Land Surface Scheme, CLASS (used in Canada into the Canadian Regional Climate Model, and the Global Climate Model), simulates the spatial and temporal evolution of snow state variables including the albedo. The albedo is computed according to the depth of snow on the ground as well as the accumulation of snow in trees. The albedo seasonal evolution for snow on ground is estimated in CLASS from an empirical aging expression with time and temperature and a “refresh” based on a threshold of snowfall depth. The seasonal evolution of snow on canopy is estimated from an interception expression with trees type and snowfall density and an empirical expression for unloading rate with time. The objectives of this project are to analyse albedo behavior (simulated and measured) and to improve CLASS simulations in winter for Eastern Canada. To do so, sensitivity test were performed on prescribed parameters (parameters that are used in CLASS computation, their values are fixed, and determined empirically). Also, albedo evolution with time and meteorological conditions were analysed for grass and coniferous terrain. Finally, we tried to improve simulations by optimizing sensitive prescribed parameters for grass and coniferous terrain, and by modifying the refresh albedo value for grass terrain. First, we analysed albedo evolution and modelling biases. Grass terrain showed strong sensitivity to the precipitation rate threshold (for the albedo to refresh to its maximum value), and to the value of the albedo refresh. Both are affected by input data of precipitation rate and phase. The modification of precipitation threshold rate generates daily surface albedo to vary mainly (75 % of data in winter) between 0.62 and 0.8, which is a greater fluctuation than for a normal simulation over winter. The modification of the albedo refresh value generates surface albedo to vary mainly (75 %) between 0.66 and 0.79, but with extreme values, 25 % of data, from 0.48 to 0.9. Coniferous areas showed small sensitivity to studied prescribed parameters. Also, comparisons were made between simulated and measured mean albedo during winter. CLASS underestimates the albedo by -0.032 (4.3 %) at SIRENE (grass in Southern Quebec), by -0.027 (3.4 %) at Goose Bay (grass in arctic site) and by -0.075 (27.1 %) at James Bay (boreal forest) (or -0.011 (5.2 %) compared to MODIS (MODerate resolution Imaging Spectroradiometer) data). A modelling issue in grass terrain is the small and steady maximum albedo value (0.84) compared to measured data in arctic condition (0.896 with variation of an order of 0.09 at Goose Bay, or 0.826 at SIRENE with warmer temperatures). In forested areas, a modelling issue is the small albedo increase (+0.17 in the visible range, +0.04 in NIR) for the part of the vegetation that is covered by snow (total surface albedo gets to a maximum of 0.22) compared to events of high surface albedo (0.4). Another bias comes from the albedo value of the snow trapped on canopy which does not decrease with time in opposition to observed surface albedo which is lower at the end of winter and which suggests snow metamorphism occurred. Secondly, we tried to improve simulations by optimizing prescribed parameters and by modifying the albedo’s maximum value computation. Optimisations were made on sensitive prescribed parameters or on those that seemed unsuited. No significant RMSE (Root Mean Square Error) improvements were obtained from optimisations in both grass and coniferous area. Improvements of albedo simulations were tried by adjusting the maximum value (normally fixed) with temperature and precipitation rate, in grass terrain. Results show that these modifications did not significantly improved simulations’ RMSE. Nevertheless, the latter modification improved the correlation between simulated and measured albedo. These statistics were made with the whole dataset which can reduce the impact of modifications (they were mainly affecting albedo during a precipitation event), but it allows to overview the new model performance. Modifications also added variability to maximum values (closer to observed albedo) and they increased our knowledge on surface albedo behavior (simulated and measured). The methodology is also replicable for other studies that would aim to analyse and improve simulations of a surface model.
Papa, Fabrice. "Nouvelles applications scientifiques des missions altimétriques pour l'étude des océans et des terres émergées." Toulouse 3, 2003. http://www.theses.fr/2003TOU30093.
Повний текст джерелаCharrois, Luc. "Assimilation de réflectances satellitaires du domaine visible et proche infrarouge dans un modèle détaillé de manteau neigeux." Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAU001/document.
Повний текст джерелаAn accurate seasonal snowpack modeling is needed to study its evolution and to improvethe avalanche hazard forecast. For 20 years, the snow study center (CEN) has developed asnowpack model named Crocus to simulate the snow cover and its physical properties drivenby near-surface meteorological conditions. Model and meteorological forcing errors are themain uncertainties in the Crocus forecasts. Constraining the model with observations canminimize the impacts of these uncertainties on simulations. Because of the low density ofground-based measurement networks combined to the high spatial variability of the snowcover, satellite observations should be the best way to constrain the model. The MODISspectroradiometer which provides daily surface information at 250 m spatial resolution isappropriated to study the snow cover. The visible and near-infrared reflectances (definedas the fraction of incident solar flux that is reflected by the surface) measured by MODISare strongly sensitive to physical properties of the snowpack. The radiative transfer modelTARTES, recently implemented into Crocus, calculates the same spectral reflectances and so,opens routes to data assimilation of MODIS reflectances.The aim of this thesis is to investigates the assimilation of the MODIS reflectances into thesnowpack model Crocus in an operational perspective. This work benefits from the expertisein physical and radiative snowpack modeling as well as data assimilation from two laboratoriesof Grenoble, the snow study center and the Laboratory of Glaciology and Geophysics of theEnvironment.The project took place in two steps to answer the following questions:Do MODIS reflectances offer an informative content allowing an efficient constraint ofthe Crocus snowpack model?What are the challenges associated to the assimilation of remotely-based optical reflectances?A particle filter is used as data assimilation scheme to evaluate the ability of opticalreflectance data assimilation to improve snow depth and snow water equivalent simulations.The choice of this filter, allowed by the small size of the problem, is based on its ease ofimplementation regarding the severe constraints of the Crocus model. The experiments wereconducted at the Col du Lautaret and the Col de Porte in the French Alps.The assimilation of synthetic observations demonstrates the potential of spectral reflectancesto constraint the Crocus snowpack model simulations. The root-mean square errors(RMSE) of bulk variables like snow depth and snow water equivalent are reduced by a factorof roughly 2 after assimilation. However, the performance of assimilation is highly dependenton the temporal distribution of the observations.The assimilation of real reflectances shows a high sensitivity to the quality of the assimilatedobservations. Converting MODIS top of atmosphere reflectances into surface reflectancesintroduces uncertainties in these data. Resulting biases and a poor characterization of errorsdeteriorate the estimation of the snowpack. Screening methods prior assimilation are thereforea priority in the prospect of satellite data assimilation.This work demonstrates the potential of remotely-based data assimilation to monitor and forecast the snow cover, potential which should be used in the near future
Mondet, Jean. "Etude des paramètres de surface de la calotte polaire antarctique, dans les domaines spectraux du visible et du proche infrarouge, à partir des données de l'instrument de télédétection POLDER." Phd thesis, Grenoble 1, 1999. http://tel.archives-ouvertes.fr/tel-00766029.
Повний текст джерелаMontpetit, Benoît. "Analyse de la modélisation de l'émission multi-fréquences micro-onde des sols et de la neige, incluant les croutes de glace à l'aide du modèle Microwave Emission Model of Layered Snowpacks (MEMLS)." Thèse, Université de Sherbrooke, 2015. http://hdl.handle.net/11143/6844.
Повний текст джерелаAbstract : Snow cover studies are essential to better understand climatic and hydrologic processes. With recent climate change observed in the northern hemisphere, more frequent rain-on-snow and meltrefreeze events have been reported, which affect the habits of the northern comunities and the survival of arctique wildlife. Passive microwave remote sensing has proven to be a great tool to characterize the state of snow cover. Nonetheless, proper modeling of the microwave signal is needed in order to understand how the parameters of the snowpack affect the measured signal. The main objective of this study is to analyze the soil, snow and ice radiative transfer in order to better characterize snow cover properties and develop an ice lens detection index with satellite passive microwave brightness temperatures. To do so, the passive microwave radiative transfer modeling of the Microwave Emission Model of Layered Snowpacks (MEMLS) was improved in order to minimize the errors on the brightness temperature simulations in the presence of ice lenses. The first improvement to passive microwave radiative transfer modeling of snow made was the snow grain size parameterization. Two new instruments, based on short wave infrared reflectance to measure the snow specific surface area (SSA) were developed. This parameter was shown to be a more accurate and objective to characterize snow grain size. The instruments showed an uncertainty of 10% to measure the SSA of snow. Also, the SSA of snow was calibrated for passive microwave modeling in order to reduce the errors on the simulated brightness temperatures. It was showed that a correction factor of φ = 1.3 needed to be applied to the grain size parameter of MEMLS, obtain through the SSA measurements, to minimize the simulation error. The second improvement to passive microwave radiative transfer modeling was the estimation of passive microwave soil emission. In-situ microwave measurements and physical temperature profiles of frozen organic arctic soils were acquired and characterized to improve the modeling of the soil emission. Effective permittivities at 10.7, 19 and 37 GHz and effective surface roughness were determined for this type of soil and the soil brightness temperature simulations were obtain with a minimal root mean square error (RMSE) of 4.65K. With the snow grain size and soil contributions to the emitted brightness temperature optimized, it was then possible to implement a passive microwave radiative transfer module of ice into MEMLS. With this module, it was possible to demonstrate that the improved Born approximation already implemented in MEMLS was equivalent to simulating a pure ice lens when the density of the layer was set to 917 kg m−3 and the grain size to 0 mm. This study also showed that by simulating ice lenses within the snow with there measured properties, the RMSE of the simulations (RMSE= 11.3 K) was similar to the RMSE for simulations of snowpacks where no ice lenses were measured (only snow, RMSE= 11.5 K). With the validated MEMLS model for snowpacks with ice lenses, an ice index was created. It is shown here that the polarization ratio (PR) was strongly affected by the presence of ice lenses within the snowpack. With simulations of the PR at 10.7, 19 and 37 GHz from measured snowpack properties in Chucrhill (Manitoba, Canada), thresholds between the measured PR and the mean winter PR were determined to detect the presence of ice within the snowpack. These thresholds were applied to a timeseries of nearly 34 years for a pixel in Nunavik (Quebec, Canada) where the soil surface is similar to that of the Churchill site. Many ice lenses are detected since 1995 with these thresholds and the same events as Roy (2014) were detected. With in-situ validation, it would be possible to confirm the precision of these thresholds but Roy (2014) showed that these events can not be explained by anything else than the presence of an ice layer within the snowpack. The same thresholds were applied to a pixel on Banks island (North-West Territories, Canada). The 2003 event that was reported by Grenfell et Putkonen (2008) was detected by the thresholds. Other events in the years 1990 and 2000’s were detected with these thresholds. These events all follow periods where the air temperature were warm and were followed by a quick drop in air temperature which could be used to validate the presence of ice layer within the snowpack. Nonetheless, without in-situ validation, these events can not be confirmed.
Allard, Matthieu. "Analyse spatio-temporelle de l'évolution des marais à scirpe de l'habitat migratoire de la Grande Oie des neiges à l'aide de l'imagerie IKONOS et de photographies aériennes." Mémoire, Université de Sherbrooke, 2008. http://savoirs.usherbrooke.ca/handle/11143/2555.
Повний текст джерелаBernard, Eric. "Les dynamiques spatio-temporelles d'un petit hydrosystème arctique : approche nivo-glaciologique dans un contexte de changement climatique contemporain (bassin du glacier Austre Lovén, Spitsberg, 79°N)." Phd thesis, Université de Franche-Comté, 2011. http://tel.archives-ouvertes.fr/tel-00910122.
Повний текст джерелаSherjal, Isabelle. "Radiométrie micro-onde de la neige : interprétation de données satellitaires sur l'Antarctique : expérimentations dans les Alpes." Phd thesis, 1995. http://tel.archives-ouvertes.fr/tel-00761401.
Повний текст джерелаTeasdale, Mylène. "Développement d’un modèle de classification probabiliste pour la cartographie du couvert nival dans les bassins versants d’Hydro-Québec à l’aide de données de micro-ondes passives." Thèse, 2015. http://hdl.handle.net/1866/12577.
Повний текст джерелаEvery day, decisions must be made about the amount of hydroelectricity produced in Quebec. These decisions are based on the prediction of water inflow in watersheds based on hydrological models. These models take into account several factors, including the presence or absence of snow. This information is critical during the spring melt to anticipate future flows, since between 30 and 40 % of the flood volume may come from the melting of the snow cover. It is therefore necessary for forecasters to be able to monitor on a daily basis the snow cover to adjust their expectations about the melting phenomenon. Some methods to map snow on the ground are currently used at the Institut de recherche d'Hydro-Québec (IREQ), but they have some shortcomings. This master thesis's main goal is to use remote sensing passive microwave data (the vertically polarized brightness temperature gradient ratio (GTV)) with a statistical approach to produce snow maps and to quantify the classification uncertainty. In order to do this, the GTV has been used to calculate a daily probability of snow via a Gaussian mixture model using Bayesian statistics. Subsequently, these probabilities were modeled using linear regression models on logits and snow cover maps were produced. The models results were validated qualitatively and quantitatively, and their integration at Hydro-Québec was discussed.