In the spring of 1999 NASA will launch the scatterometer SeaWinds, beginning a 3 year mission to measure the ocean winds. SeaWinds is different from previous spaceborne scatterometers in that it employs a rotating pencil-beam antenna as opposed to fixed fan-beam antennas. The scanning beam provides greater coverage but causes the wind retrieval accuracy to vary across the swath. This thesis develops a filed-wise wind retrieval algorithm to improve the overall wind retrieval accuracy for use with SeaWinds data. In order to test the field-wise wind retrieval algorithm, methods for simulating wind fields are developed. A realistic approach interpolates the NASA Scatterometer (NSCAT) estimates to fill a SeaWinds swath using optimal interpolation along with linear wind filed models. The two stages of the field-wise wind retrieval algorithm are filed-wise estimation and field-wise ambiguity selection. Field-wise estimation is implemented using a 22 parameter Karhunen-Loeve (KL) wind field model in conjunction with a maximum likelihood objective function. An augmented multi-start global optimization is developed which uses information from the point-wise estimates to aid in a global search of the objective function. The local minima in the objective function are located using the augmented multi-start search techniques and are stored as field-wise ambiguities. The ambiguity selection algorithm uses a field-wise median filter to select the field-wise ambiguity closest to the true wind in each region. Point-wise nudging is used to further improve the filed-wise estimate using information from the point-wise estimates. Combined, these two techniques select a good estimate of the wind 95% of the time. The overall performance of the field-wise wind retrieval algorithm is compared with the performance of the current point-wise techniques. Field-wise estimation techniques are shown to be potentially better than point-wise techniques. The field-wise estimates are also shown to be very useful tools in point-wise ambiguity selection since 95.8%-96.6% of the point-wise estimates closest to the field-wise estimates are the correct aliases.

Avec une réduction de la taille des composants en constante progression, le domaine de la microélectronique, et d'une manière plus globale, le domaine de la nanofabrication se doit de posséder des outils de métrologie dimensionnelle performants. L'amélioration de points pertinents comme la rapidité, la précision et la répétabilité devrait permettre un suivi en temps réel de l'évolution des procédés et ainsi améliorer les rendements de production tout en limitant les pertes imputables aux dérives des procédés. Dans ce cadre, la scattérométrie, technique optique de métrologie dimensionnelle basée sur l'analyse de la lumière diffractée, a montré, suivant les cas, des capacités à répondre aux exigences des applications temps réel. Elle se décompose en une phase de mesure, effectuée par un dispositif expérimental (ellipsomètre dans notre cas) et une phase de résolution de problème inverse. La méthode utilisée pour traiter cette dernière phase conditionne la compatibilité avec le temps réel. La méthode des bibliothèques et une méthode utilisant des réseaux de neurones artificiels présentent les qualités requises. La première a déjà été validée pour le suivi d'un procédé de gravure en microélectronique et la seconde a été testée uniquement en statique à la suite d'une étape technologique. Cette thèse a pour but d'évaluer l'apport des réseaux de neurones en scattérométrie dynamique. Basée sur des critères qualitatifs et quantitatifs, cette étude souligne également la difficulté de comparer avec objectivité les différentes techniques de métrologie. Ces travaux dressent également une comparaison minutieuse de ces deux méthodes adaptées au temps réel afin d'en dégager les spécificités de fonctionnement. Enfin, la scattérométrie par l'approche des réseaux de neurones est étudiée dans le cas de la gravure de résine par plasma. En effet, il s'agit d'un procédé de fabrication en microélectronique pour lequel le contrôle in-situ est un enjeu important dans le futur. / The decrease of the components size has been widely witnessed in the past decades. Hence, microelectronic field, and more generally speaking, nanofabrication requires very efficient dimensional metrology tools. The improvement of relevant points like the speed, the accuracy and the repeatability of the tool will allow real time process monitoring and thus enhance the production yield while restricting the waste due to process drift. In this framework, scatterometry, an optical dimensional metrology technique based on the analysis of the diffracted light, has proven its ability to meet real time applications requirements. It is composed of a measuring phase, done by an experimental setup (ellipsometer in our case) and an inverse problem resolution phase. The chosen method used in order to process this last step determines the compatibility with real time. Library method and a method based on artificial neural networks possess the required qualifications. The first one has already been validated for etching process monitoring in microelectronics and the second one has been validated only on static cases after a technological step. This PhD involves assessing neural networks for dynamic scatterometry. Based on qualitative and quantitative criteria, this study underlines the difficulty of comparing different metrology techniques objectively. This work draws up a meticulous comparison of these two real time adapted methods in order to bring out their working specifications. Finally, scatterometry using neural networks is studied on a resist etching plasma case. In fact, this is a microelectronic fabrication process for which in-situ control is of an important concern in the future.

Oceanic microwave remote sensing provides the data necessary for the estimation of significant geophysical parameters such as the near-surface vector wind. To obtain accurate estimates, a precise understanding of the measurements is critical. This work clarifies and quantifies specific uncertainties in the scattered power measured by an active radar instrument. While there are many sources of uncertainty in remote sensing measurements, this work concentrates on three significant, yet largely unstudied effects. With a theoretical derivation of the backscatter from an ocean-like surface, results from this dissertation demonstrate that the backscatter decays with surface roughness with two distinct modes of behavior, affected by the size of the footprint. A technique is developed and scatterometer data analyzed to quantify the variability of spaceborne backscatter measurements for given wind conditions; the impact on wind retrieval is described in terms of bias and the Cramer-Rao lower bound. The probability density function of modified periodogram averages (a spectral estimation technique) is derived in generality and for the specific case of power estimates made by the NASA scatterometer. The impact on wind retrieval is quantified.

Because the earth's cryosphere influences global weather patterns and climate, the scientific community has had great interest in monitoring this important region. Microwave remote sensing has proven to be a useful tool in estimating sea and glacial ice surface characteristics with both scatterometers and radiometers exhibiting high sensitivity to important ice properties. This dissertation presents an array of studies focused on extracting key surface features from multisensor microwave data sets. First, several enhanced resolution image reconstruction issues are addressed. Among these are the optimization of the scatterometer image reconstruction (SIR) algorithm for NASA scatterometer (NSCAT) data, an analysis of Ku-band azimuthal modulation in Antarctica, and inter-sensor European Remote Sensing Satellite (ERS) calibration. Next, various methods for the removal of atmospheric distortions in image reconstruction of passive radiometer observations are considered. An automated algorithm is proposed which determines the spatial extent of sea ice in the Arctic and Antarctic regions from NSCAT data. A multisensor iterative sea ice statistical classification method which adapts to the temporally varying signatures of ice types is developed. The sea ice extent and classification algorithms are adopted for current SeaWinds scatterometer data sets. Finally, the automated inversion of large-scale forward electromagnetic scattering of models is considered and used to study the temporal evolution of the scattering properties of polar sea ice.

The Ku band microwave remote sensor, SeaWinds, was developed at the National Aeronautics and Space Administration (NASA) Jet Propulsion Laboratory (JPL). Two identical SeaWinds instruments were launched into space. The first was flown onboard NASA QuikSCAT satellite which has been orbiting the Earth since June 1999, and the second instrument flew onboard the Japanese Advanced Earth Observing Satellite II (ADEOS-II) from December 2002 till October 2003 when an irrecoverable solar panel failure caused a premature end to the ADEOS-II satellite mission. SeaWinds operates at a frequency of 13.4 GHz, and was originally designed to measure the speed and direction of the ocean surface wind vector by relating the normalized radar backscatter measurements to the near surface wind vector through a geophysical model function (GMF). In addition to the backscatter measurement capability, SeaWinds simultaneously measures the polarized radiometric emission from the surface and atmosphere, utilizing a ground signal processing algorithm known as the QuikSCAT / SeaWinds Radiometer (QRad / SRad). This dissertation presents the development and validation of a mathematical inversion algorithm that combines the simultaneous active radar backscatter and the passive microwave brightness temperatures observed by the SeaWinds sensor to retrieve the oceanic rainfall. The retrieval algorithm is statistically based, and has been developed using collocated measurements from SeaWinds, the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) rain rates, and Numerical Weather Prediction (NWP) wind fields from the National Centers for Environmental Prediction (NCEP). The oceanic rain is retrieved on a spacecraft wind vector cell (WVC) measurement grid that has a spatial resolution of 25 km. To evaluate the accuracy of the retrievals, examples of the passive-only, as well as the combined active / passive rain estimates from SeaWinds are presented, and comparisons are made with the standard TRMM rain data products. Results demonstrate that SeaWinds rain measurements are in good agreement with the independent microwave rain observations obtained from TMI. Further, by applying a threshold on the retrieved rain rates, SeaWinds rain estimates can be utilized as a rain flag. In order to evaluate the performance of the SeaWinds flag, comparisons are made with the Impact based Multidimensional Histogram (IMUDH) rain flag developed by JPL. Results emphasize the powerful rain detection capabilities of the SeaWinds retrieval algorithm. Due to its broad swath coverage, SeaWinds affords additional independent sampling of the oceanic rainfall, which may contribute to the future NASA's Precipitation Measurement Mission (PMM) objectives of improving the global sampling of oceanic rain within 3 hour windows. Also, since SeaWinds is the only sensor onboard QuikSCAT, the SeaWinds rain estimates can be used to improve the flagging of rain-contaminated oceanic wind vector retrievals. The passive-only rainfall retrieval algorithm (QRad / SRad) has been implemented by JPL as part of the level 2B (L2B) science data product, and can be obtained from the Physical Oceanography Distributed Data Archive (PO.DAAC).

Cette thèse s'intéresse à la dynamique océanique induite par les échelles spatiales et temporelles de l'atmosphère, et du vent en particulier, dans les régions d'upwelling du Benguela et des Canaries. Ces régions sont sous l'influence d'un vent local ou régional, soufflant parallèlement à la côte. "Moteur" principal de la résurgence d'eau froide, ce vent est modulé par des processus physiques à des échelles spatio-temporelles variées. La nature des interactions avec l'atmosphère, l'océan et le continent environnants diffère selon les processus. Depuis deux décennies, des efforts remarquables portent sur la description par télédétection des champs atmosphériques à la surface de l’océan. Un nombre croissant de missions spatiales et des améliorations techniques majeures ont permis de raffiner la résolution horizontale et temporelle des produits disponibles à l'échelle globale. La disponibilité de multiples mesures diffusiométriques grillées, traitées et distribuées par le LOSCERSAT,nous amène dans un premier temps à comparer et analyser la richesse et la finesse des échelles retranscrites par différents produits. Ainsi, plusieurs gammes d'échelles de vent sont différenciées et leurs signatures sur l'upwelling côtier sont étudiées. L'intensité des anticyclones subtropicaux (Sainte Hélène et Açores) module la saisonnalité de l'upwelling le long des côtes Ouest africaines. Les régions centrales des upwellings de l’Atlantique, sous l'influence permanente de ces centres de haute pression, sont ainsi les cellules d'upwelling les plus intenses de chacun des systèmes en termes de pérennité et d'intensité (cellule de Lüderitz et cellule de Dakhla respectivement dans l'hémisphère Sud et l'hémisphère Nord). À l'échelle régionale, ou l'échelle des sous-bassins (O(1000 km)), la variabilité intrasaisonnière du vent est contrôlée par le renforcement ou l'atténuation des anticyclones entraînant à la côte l'activation ou la relaxation d'événements d'upwelling. À des échelles plus petites (O(100 km)), le front caractéristique de température de surface (SST) entre la côte et le large façonne la structure spatiale du vent par des processus de stabilisation/déstabilisation de la colonne d'air. Un vent soufflant en direction de l'équateur et parallèlement à un front de SST aura tendance à diminuer (augmenter) sur le flanc froid (chaud) de ce front. Le rotationnel (la divergence) du vent est directement impacté(e) et répond linéairement, au premier ordre, à la composante du gradient de SST normale (tangentielle) à la direction du vent. Ces rétroactions océaniques sont caractérisées par une échelle temporelle allant de l'hebdomadaire au mensuel. Enfin, de fines échelles du vent sont couramment observées dans les premiers kilomètres de l’océan au voisinage de la côte. L’interface entre le large et le continent est en effet associée à un affaiblissement significatif des vents. L'extension zonale de cette transition (O(10 km)) dépend notamment de l'orographie et de la rugosité de surface du continent adjacent. L'impact d'une telle réduction du vent sur la structure des upwellings côtiers, la dynamique sous-jacente et le transport côte-large de particules est appréhendé à l'aide d'analyses numériques eulériennes et lagrangiennes. / This study focuses on the oceanic response to fine atmospheric spatial and temporal scales, and especially fine wind patterns in the Benguela and Canary upwelling systems. These regions are under the influence of local or regional wind, blowing parallel to the coast. Thewind is the main driver of the cold-water upwelling and is modulated by several physical processes at various scales. The nature of the interactions with the atmosphere, the ocean and the adjacent continent differs according to these processes. For the past 20 years, outstanding efforts have been made in the description and understanding of the atmospheric conditions at the sea surface. An increasing number of space missions and major technical improvements have allowed refinement of the horizontaland temporal resolution of the products available at global scale. The availability of multiple gridded scatterometer measurements,processed and distributed by the LOS-CERSAT, brings us first to compare and analyze the richness and fineness of the scales of a few products. We differentiate several wind scales and study their signatures on coastal upwelling dynamics. The intensity of the subtropical anticyclones (Saint Helena andAzores) modulates the seasonality of the upwelling along the Africanwest coast. The central regions of both upwelling systems are permanently under the influence of these atmospheric highs and,thus, are the most intense upwelling cells of each system, both interms of durability and intensity (Lüderitz and Dakhla cells for the southern and the northern hemisphere, respectively). On a regional scale, or basin scale (O(1000 km)), the intraseasonal wind variability is driven by the strengthening or weakening of these anticyclones, causing the activation or relaxation of upwelling events at the coast.At smaller scales (O(100 km)), the characteristic sea surface temperature (SST) front between the coastal and open ocean shapes the spatial structure of the wind by stabilization/destabilization of the air column. An equatorward-blowing wind parallel to an SST front tends to decrease (increase) on the cold side (warm) of this front. The curl (divergence) of the wind is directly impacted and the first order response varies linearly with the crosswind (downwind) SST gradient. This oceanic feedback is characterized by weekly to monthly temporal scales. Finally, small-scale wind structures are frequently observed in the first kilometers of the coastal ocean. Indeed, the interface between the open ocean and the continent is associated with a significant wind drop-off. The zonal extension of this transition (O(10 km) depends on the orography and on the surface roughness of the adjacent continent. The impact of such a wind reduction on the structure of the coastal upwelling, the underlying ocean dynamics and the cross-shore transport of particles is diagnosed with both Eulerian and Lagrangian numerical analyses.