Microwave radar and radiometer techniques are used to gather crucial information about the earth and its atmosphere. The ERS-1, JERS-1, RadarSAT and NASA’s Mission to Planet Earth projects are designed to study the changing global environment. In all these endeavors, the key instrument is the radar or scatterometer. The advantage of microwave radar is that it is hindered very little by clouds, fog or solar radiation. Polarimetrie sensors like the shuttle-borne SIR-C radar, provides additional information compared to single polarization systems. Correct interpretation of polarimetrie data necessitates proper understanding of the scattering mechanism. Thus theory of polarization synthesis is discussed. Solution to the Kennaugh eigenvalue problem for point targets is derived. Polarimetrie signatures of point targets are shown as surfaces of spherical co-ordinates based on the Poincare sphere. Statistics of the covariance matrix elements for distributed targets are presented. The main topic of this research report is the design and development of a multifrequency, polarimetrie scatterometer for biosphere remote sensing. The system was developed using a standard HP network analyzer, a crossed log-periodic dipole antenna and a reflector. The scatterometer functions in a linear polarization basis between the L- and X-bands and gathers full-polarimetric information. The standard S-parameter measurements using the network analyzer were related to surface and volume scattering coefficients of rough surface, snow cover and vegetation media. The scatterometer measurements were carried out in the frequency domain to make use of narrow band filters in the receiver chain. The fast Fourier transform was used to convert the frequency domain measurements to the time domain. The range resolution of the system was 20 cm; azimuthal and elevation resolutions are determined by the antenna beam widths. Range side lobes were reduced by making use of appropriate weighting (Kaiser-Bessel window) functions. In the process of receiver design, we developed a number of signal processing techniques which are illustrated using appropriate numerical examples. The accuracy of target characterization depends on the quality of scatterometer calibration. A novel technique to estimate the absolute gain and crosstalk of the radar system was developed. Using a distortion matrix approach, the cross-polarization response of the system was improved by 10 to 25 dB. The radar measurements were validated by comparing point target radar observations with the corresponding theoretical values. Also, measurements of fading decorrelation distance and decorrelation bandwidth of rough surfaces were in good agreement with the theory. Backscatter observations of vegetation and snow cover were comparable to earlier published values for a similar environment. Based on initial test results and operations capability, we propose to use the present scatterometer for ground-truthing in support of ERS-1 missions. Direct comparisons of electromagnetic backscatter coefficients are possible between the ERS-1 and the present scatterometer. These joint studies are beneficial for developing inverse scattering techniques, designing new experiments and calibrating ERS-1 radar systems for distributed target environments. / <p>Diss. Umeå : Umeå universitet, 1995</p> / digitalisering@umu

Thesis (Ph.D.)--Electrical and Computer Engineering, Georgia Institute of Technology, 2008. / Committee Chair: Dr. Paul G. Steffes; Committee Member: Dr. Gregory D. Durgin; Committee Member: Dr. Robert D. Braun; Committee Member: Dr. Thomas K. Gaylord; Committee Member: Dr. Waymond R. Scott

La région de l'archipel canadien, située en Arctique, connaît actuellement d'importants changements climatiques, se traduisant notamment par une augmentation des températures, une réduction de l'étendue de la banquise marine et du couvert nival terrestre ou encore une perte de masse significative des calottes glaciaires disséminées sur les îles de l'archipel. Parmi ces calottes glaciaires, la calotte Barnes, située en Terre de Baffin, ne fait pas exception comme le montrent les observations satellitaires qui témoignent d'une importante perte de masse ainsi que d'une régression de ses marges, sur les dernières décennies. Bien que les calottes glaciaires de l'archipel canadien ne représentent que quelques dizaines de centimètres d'élévation potentielle du niveau des mers, leur perte de masse est une composante non négligeable de l'augmentation actuelle du niveau des mers. Les projections climatiques laissent à penser que cette contribution pourrait rester significative dans les décennies à venir. Cependant, afin d'estimer les évolutions futures de ces calottes glaciaires et leur impact sur le climat ou le niveau des mers, ils est nécessaire de caractériser les processus physiques tels que les modifications du bilan de masse de surface. Cette connaissance est actuellement très limitée du fait notamment du sous-échantillonnage des régions arctiques en terme de stations météorologiques permanentes. Une autre particularité de certaines calottes de l'archipel canadien, et de la calotte Barnes en particulier, est de présenter un processus d'accumulation de type glace surimposée, ce phénomène étant à prendre en compte dans l'étude des processus de surface. Pour palier au manque de données, l'approche retenue a été d'utiliser des données de télédétection, qui offrent l'avantage d'une couverture spatiale globale ainsi qu'une bonne répétitivité temporelle. En particulier les données acquises dans le domaine des micro-ondes passives est d'un grand intérêt pour l'étude de surfaces enneigées. En complément de ces données, la modélisation du manteau neigeux, tant d'un point de vue des processus physiques que de l'émission électromagnétique permet d'avoir accès à une compréhension fine des processus de surface tels que l'accumulation de la neige, la fonte, les transferts d'énergie et de matière à la surface, etc. Ces différents termes sont regroupés sous la notion de bilan de masse de surface. L'ensemble du travail présenté dans ce manuscrit a donc consisté à développer des outils permettant d'améliorer la connaissance des processus de surface des calottes glaciaires du type de celles que l'on rencontre dans l'archipel canadien, l'ensemble du développement méthodologique ayant été réalisé sur la calotte Barnes à l'aide du schéma de surface SURFEX-CROCUS pour la modélisation physique et du modèle DMRT-ML pour la partie électromagnétique. Les résultats ont tout d'abord permis de mettre en évidence une augmentation significative de la durée de fonte de surface sur la calotte Barnes (augmentation de plus de 30% sur la période 1979-2010), mais aussi sur la calotte Penny, elle aussi située en Terre de Baffin et qui présente la même tendance (augmentation de l'ordre de 50% sur la même période). Ensuite, l'application d'une chaîne de modélisation physique contrainte par diverses données de télédétection a permis de modéliser de manière réaliste le bilan de masse de surface de la dernière décennie, qui est de +6,8 cm/an en moyenne sur la zone sommitale de la calotte, qui est une zone d'accumulation. Enfin, des tests de sensibilité climatique sur ce bilan de masse ont permis de mettre en évidence un seuil à partir duquel cette calotte voit disparaître sa zone d'accumulation. Les modélisations effectuées suggèrent que ce seuil a de fortes chances d'être atteint très prochainement, pour une augmentation de température moyenne inférieure à 1°C, ce qui aurait pour conséquence une accélération de la perte de masse de la calotte. / Significant climate change is curently monitored in the Arctic, and especially in the region of the canadian arctic archipellago. This climate warming leads to recession of sea-ice extent and seasonnal snow cover, and also to large mass loss of the archipellago's ice caps. One of the most southern ice cap, the Barnes Ice Cap, located on the Baffin Island, is no exception to significant mass loss and margins recession as satellite observations exhibited over the last decades. Despite the relative low sea level potential of the small ice caps located in the canadian arctic achipellago in regards to major ice sheets, Antarctica and Greenland, their contribution to the current sea level rise is significant. Climate projections show that this contribution could accelerate significant over the next decades. However, to estimate the future evolution of these ice caps and their impact on climate or sea level rise, a better characterisation of the surface processes such as the evolution of the surface mass balance is needed. This knowledge is currently very limited, mainly due to the sparse covering of automatic weather stations or in-situ measurements over the Arctic. Furthermore, several ice caps, among with the Barnes Ice Cap, present a superimposed ice accumulation area which particularities have to be taken into account in the surface processes studies. Given the lack of in-situ data, the approach choosen in this work is to use remote sensing data, that have the advantage to offer a good spatial and temporal coverage. In particular, passive microwave data are very suitable for snowy surfaces studies. To complement these data, physical and electromagnetic snowpack modeling provide a fine characterisation of surface processes such as snow accumulation. The whole work presented in this manuscript thus consisted in developping specific tools to improve the understanding of surface processes of small arctic ice caps. This methodological development was performed and applied on the Barnes Ice Cap using the surface scheme SURFEX-CROCUS and the electromagnetic model DMRT-ML. First results highlight a significant increase in surface melt duration over the past 3 decades on the Barnes Ice Cap (increase of more than 30% over 1979-2010 period). A similar trend is also monitored over the Penny Ice Cap, located in the south part of the Baffin Island (increase of more than 50% over the same period). Then, the surface mass balance over the last decade was modeled by using a physical based modeling chain constrained by remote sensing data. The results give a mean net accumulation of +6,8 cm/an on the summit area of the ice cap. Finaly, sensitivity tests, performed to investigate the climatic sensitivity of the surface mass balance, highlight a threshold effect that may lead to a complete disapearence of the accumulation area of the Barnes Ice Cap. With a temperature increase less than 1°C, modeling results suggest it is likely that the threshold will be reached rapidly leading to an increase in mass loss from the ice cap.

The Method of Ordered Multiple Interactions provides a numerical solution to the integral<br />equations describing surface scattering which is both computationally efficient and reliably<br />convergent. The method has been applied in a variety of ways to solving the electromagnetic<br />scattering from perfectly-conducting rough surfaces. A desire to more accurately predict<br />the scattering from natural terrain has led to the representation of the surface material as<br />penetrable instead of conductive.<br /><br />For this purpose, the Method of Ordered Multiple Interactions is applied to numerically<br />solve the electromagnetic scattering from randomly-rough dielectric surfaces. A primary<br />consequence of the penetrable surface material is the introduction of a pair of coupled integral equations in place of the single integral equation used to solve the problem with a perfectly conducting surface. The method is tested and analyzed by developing independent scattering solutions for canonical cases in a transform domain and by comparing results with solutions from other techniques.<br /><br />The dielectric implementation of the Method of Ordered Multiple Interactions is used to solve<br />the electromagnetic scattering from a class of randomly-rough dielectric surfaces. This allows<br />for the characterization of the effect of a number of transmitter and surface parameters in the<br />scattering problem, observing bistatically and also specifically in the backscatter direction.<br /><br />MOMI is then applied as a method to examine subsurface penetration characteristics from<br />a similar family of rough surfaces. Characteristics of the environment parameters and the<br />scattered field itself are examined, and the numerical challenges associated with observing<br />beneath the surface are identified and addressed.<br /><br />The Method of Ordered Multiple Interactions is then incorporated as a major component of<br />a larger solution which computes the total scattering when a dielectric object is buried just<br />beneath the rough surface. This hyrid approach uses MOMI and the Method of Moments to<br />iteratively account for multiple interactions between the target and the dielectric interface,<br />enabling the study of scattering from the combined environment of a rough surface and the<br />embedded object, as well as the individual scattering events which combine to form the<br />steady-state solution. / Ph. D.

A 94.8 GHz interferometric imaging system utilizing aperture synthesis and tomography is developed for the Center for Advanced Sensor and Communication Antennas. Whereas typical interferometer designs employ multiple antennas to synthesize an aperture for image reconstruction, this unique interferometer will reproduce a scene's brightness temperature with only two antennas. To achieve this, the aperture synthesis is done with one antenna remaining stationary while the second antenna is moved at discrete increments along two controlled tracks. The two signals received by the antennas are cross-correlated to produce measured visibility function samples. The visibility samples reconstruct the scene brightness temperature through an inverse Fourier transform relationship.

Long term trends in Arctic sea ice are of particular interest with regard to global temperature, climate change, and industry. This thesis uses microwave scatterometer data from QuikSCAT and radiometer data to analyze intra- and interannual trends in first-year and multi-year Arctic sea ice. It develops a sea ice type classification method. The backscatter of first-year and multi-year sea ice are clearly identifiable and are observed to vary seasonally. Using an average of the annual backscatter trends obtained from QuikSCAT, a classification of multi-year ice is obtained which is dependent on the day of the year (DOY). Validation of the classification method is done using regional ice charts from the Canadian Ice Service. Differences in ice classification are found to be less than 6% during the winters of 06-07, 07-08, and the end of 2008. Anomalies in the distribution of sea ice backscatter from year to year suggest a reduction in multi-year ice cover between 2003 and 2009 and an approximately equivalent increase in first-year ice cover.

Satellite borne radar scatterometers provide frequent estimates of near surface wind vectors over the Earth's oceans. However in the polar oceans, the presence of sea ice in or near the measurement footprint can adversely a ect scatterometer measurements resulting in inaccurate wind estimates. Currently, such ice contamination is mitigated by discarding measurements within 50 km of detected sea ice. This approach is imperfect and causes loss of coverage. This thesis presents a new algorithm which detects ice-contaminated measurements based on a metric called the Ice Contribution Ratio (ICR) which measures the spatial ice contribution for each measurement. The ICR calculation is made for each measurement using a spatial ice probability map which is determined using Bayesian probability theory. Determined by simulation, the ICR processing thresholds the ICR for each measurement depending on local wind, ice backscatter, and cross-track location. ICR processing retrieves winds at a distance of 22.5 km from the ice edge on average, while ensuring wind accuracy. Retrieved wind distributions using ICR processing more closely resembles uncontaminated wind distributions than winds retrieved using previous methods. The algorithm is applied to QuikSCAT in this thesis but could be applied to other scatterometers such as the Oceansat-2 scatterometer.

This research concerns the CONAE Microwave Radiometer (MWR), on board the Aquarius/SAC-D platform. MWR's main purpose is to provide measurements that are simultaneous and spatially collocated with those of NASA's Aquarius radiometer/scatterometer. For this reason, knowledge of the MWR antenna beam footprint geolocation is crucial to mission success. In particular, this thesis addresses an on-orbit validation of the MWR antenna beam pointing, using calculated MWR instantaneous field of view (IFOV) centers, provided in the CONAE L-1B science data product. This procedure compares L-1B MWR IFOV centers at land/water crossings against high-resolution coastline maps. MWR IFOV locations versus time are computed from knowledge of the satellite's instantaneous location relative to an earth-centric coordinate system (provided by on-board GPS receivers), and a priori measurements of antenna gain patterns and mounting geometry. Previous conical scanning microwave radiometer missions (e.g., SSM/I) have utilized observation of rapid change in brightness temperatures (T_B) to estimate the location of land/water boundaries, and subsequently to determine the antenna beam-pointing accuracy. In this thesis, results of an algorithm to quantify the geolocation error of MWR beam center are presented, based upon two-dimensional convolution between each beam's gain pattern and land-water transition. The analysis procedures have been applied to on-orbit datasets that represent land-water boundaries bearing specific desirable criteria, which are also detailed herein. The goal of this research is to gain a better understanding of satellite radiometer beam-pointing error and thereby to improve the geolocation accuracy for MWR science data products.

The removal of systematic brightness temperature (Tb) biases is necessary when producing decadal passive microwave data sets for weather and climate research. It is crucial to achieve Tb measurement consistency among all satellites in a constellation as well as to maintain sustained calibration accuracy over the lifetime of each satellite sensor. In-orbit inter-satellite radiometric calibration techniques provide a long term, group-wise solution; however, since radiometers operate at different frequencies and viewing angles, Tb normalizations are made before making intermediate comparisons of their near-simultaneous measurements. In this dissertation, a new approach is investigated to perform these normalizations from one satellite's measurements to another. It uses Taylor's series expansion around a source frequency to predict Tb of a desired frequency. The relationship between Tb's and frequencies are derived from simulations using an oceanic Radiative Transfer Model (RTM) over a wide variety of environmental conditions. The original RTM is built on oceanic radiative transfer theory. Refinements are made to the model by modifying and tuning algorithms for calculating sea surface emission, atmospheric emission and attenuations. Validations were performed with collocated WindSat measurements. This radiometric calibration approach is applied to establish an absolute brightness temperature reference using near-simultaneous pair-wise comparisons between a non-sun synchronous radiometer and two sun-synchronous polar-orbiting radiometers: the Tropical Rain Measurement Mission (TRMM) Microwave Imager (TMI), WindSat (on Coriolis) and Advanced Microwave Scanning Radiometer (AMSR) on Advanced Earth Observing System -II (ADEOSII), respectively. Collocated measurements between WindSat and TMI as well as between AMSR and TMI, within selected 10 weeks in 2003 for each pair, are collected, filtered and applied in the cross calibration. AMSR is calibrated to WindSat using TMI as a transfer standard. Accuracy prediction and error source analysis are discussed along with calibration results. This inter-satellite radiometric calibration approach provides technical support for NASA's Global Precipitation Mission which relies on a constellation of cooperative satellites with a variety of microwave radiometers to make global rainfall measurements.

An electromagnetic model is developed for predicting the microwave blackbody emission from the ocean surface over a wide range of frequencies, incidence angles, and wind vector (speed and direction) for both horizontal and vertical polarizations. This ocean surface emissivity model is intended to be incorporated into an oceanic radiative transfer model to be used for microwave radiometric applications including geophysical retrievals over oceans. The model development is based on a collection of published ocean emissivity measurements obtained from satellites, aircraft, field experiments, and laboratory measurements. This dissertation presents the details of methods used in the ocean surface emissivity model development and comparisons with current emissivity models and aircraft radiometric measurements in hurricanes. Especially, this empirically derived ocean emissivity model relates changes in vertical and horizontal polarized ocean microwave brightness temperature measurements over a wide range of observation frequencies and incidence angles to physical roughness changes in the ocean surface, which are the result of the air/sea interaction with surface winds. Of primary importance are the Stepped Frequency Microwave Radiometer (SFMR) brightness temperature measurements from hurricane flights and independent measurements of surface wind speed that are used to define empirical relationships between C-band (4 - 7 GHz) microwave brightness temperature and surface wind speed. By employing statistical regression techniques, we develop a physical-based ocean emissivity model with empirical coefficients that depends on geophysical parameters, such as wind speed, wind direction, sea surface temperature, and observational parameters, such as electromagnetic frequency, electromagnetic polarization, and incidence angle.