Engineering topics

Fung, A. K.

Publications and source records attributed to Fung, A. K..

At least 55 records · Page 3

NASA NTRS1986

Radiative transfer in multilayered random medium with laminar structure - Green's function approach

For a multilayered random medium with a laminar structure a Green's function approach is introduced to obtain the emitted intensity due to an arbitrary point source. It is then shown that the approach is applicable to both active and passive remote sensing. In active remote sensing, the computed radar backscattering cross section for the multilayered medium includes the effects of both volume multiple scattering and surface multiple scattering at the layer boundaries. In passive remote sensing, the brightness temperature is obtained for arbitrary temperature profiles in the layers. As an illustration the brightness temperature and reflectivity are calculated for a bounded layer and compared with results in the literature.

Karam, M. A.↗

NASA NTRS1986

Scattering from a random layer embedded with dielectric needles

Intensity scattering from a random layer imbedded with small dielectric needles is studied for applications to coniferous vegetation. The phase matrix of a thin needle whose length may be appreciable compared to the incident wavelength is presented. The effects of needle orientation on scattering is taken into account by averaging the phase function over angles of orientation. The backscattering coefficient from the layer is computed by solving the radiative transfer equation. The effects of operating frequency, orientation and size of a needle on like- and cross-backscattering are demonstrated. It was found that in backscattering angular trends are mainly controlled by the orientation of the needles.

Eom, H. J.↗

NASA NTRS1986

Microwave remote sensing: Active and passive. Volume 3 - From theory to applications

Aspects of volume scattering and emission theory are discussed, taking into account a weakly scattering medium, the Born approximation, first-order renormalization, the radiative transfer method, and the matrix-doubling method. Other topics explored are related to scatterometers and probing systems, the passive microwave sensing of the atmosphere, the passive microwave sensing of the ocean, the passive microwave sensing of land, the active microwave sensing of land, and radar remote sensing applications. Attention is given to inversion techniques, atmospheric attenuation and emission, a temperature profile retrieval from ground-based observations, mapping rainfall rates, the apparent temperature of the sea, the emission behavior of bare soil surfaces, the emission behavior of vegetation canopies, the emission behavior of snow, wind-vector radar scatterometry, radar measurements of sea ice, and the back-scattering behavior of cultural vegetation canopies.

Ulaby, F. T.↗

NASA NTRS1986

Methods for the solution of radiative transfer equation

To obtain an exact solution of the radiative-transfer equation in media where both absorption and scattering are significant, the usual approach is to use a numerical method. Three methods are known in the literature: invariant imbedding, eigenvalue-eigenfunction, and matrix doubling. This paper examines the practical application of these methods to the problem of emission from an inhomogeneous (Rayleigh) layer, the effects of layer parameters on the stability. It is found that invariant imbedding is most suitable for computing emission from an inhomogeneous layer with a temperature profile but tends to be unstable as the optical thickness of the layer increases beyond 0.5. On the other hand, the matrix-doubling method is stable for arbitrary optical thickness but is not suitable for handling multilayers. The eigenvalue-eigenfunction method is more stable than the invariant imbedding as optical thickness increases up to 2.0. It also permits temperature profile in the layer, but the computation is much more complicated. It is less stable than the matrix-doubling method when optical thickness is larger than 2.0. In general, the choice of a method is dependent on the nature of the problem.

Chen, M. F.↗

NASA NTRS1985

A study of backscattering and emission from closely packed inhomogeneous media

The effects of close spacing between small scattering spheres were examined by keeping the distance-dependent terms in the expressions for the transverse scattered fields. The phase matrix was then derived from these fields and was used in the radiative transfer formulation to model scattering and emission from a densely populated, inhomogeneous layer. Computed results were compared with those obtained when the phase matrix was specialized to the far-field condition. It was found that the use of the far-zone condition tended to underestimate both the level of the copolarized backscattering and the cross-polarized backscattering. In emission computations, the use of the far-zone condition overestimated the level of the brightness temperature. These effects decreased with a decrease in the volume fraction or an increase in the exploring frequency, as expected. An improvement on the snow parameter (density and crystal size) estimation was shown to be possible when this new phase matrix was used.

Fung, A. K.↗

NASA NTRS1985

A comparison between active and passive sensing of soil moisture from vegetated terrains

A comparison between active and passive sensing of soil moisture over vegetated areas is studied via scattering models. In active sensing three contributing terms to radar backscattering can be identified: (1) the ground surface scatter term; (2) the volume scatter term representing scattering from the vegetation layer; and (3) the surface volume scatter term accounting for scattering from both surface and volume. In emission three sources of contribution can also be identified: (1) surface emission; (2) upward volume emission from the vegetation layer; and (3) downward volume emission scattered upward by the ground surface. As ground moisture increases, terms (1) and (3) increase due to increase in permittivity in the active case. However, in passive sensing, term (1) decreases but term (3) increases for the same reason. This self compensating effect produces a loss in sensitivity to change in ground moisture. Furthermore, emission from vegetation may be larger than that from the ground. Hence, the presence of vegetation layer causes a much greater loss of sensitivity to passive than active sensing of soil moisture.

Fung, A. K.↗

NASA NTRS1985

Scattering Models and Basic Experiments in the Microwave Regime

The objectives of research over the next three years are: (1) to develop a randomly rough surface scattering model which is applicable over the entire frequency band; (2) to develop a computer simulation method and algorithm to simulate scattering from known randomly rough surfaces, Z(x,y); (3) to design and perform laboratory experiments to study geometric and physical target parameters of an inhomogeneous layer; (4) to develop scattering models for an inhomogeneous layer which accounts for near field interaction and multiple scattering in both the coherent and the incoherent scattering components; and (5) a comparison between theoretical models and measurements or numerical simulation.

Fung, A. K.↗

NASA NTRS1985

Microwave model prediction and verifications for vegetated terrain

To understand the scattering properties of a deciduous and a coniferous type vegetation scattering models were developed assuming either a disc type leaf or a needle type leaf. The major effort is to calculate the corresponding scattering phase functions and then each of the functions is used in a radiative transfer formulation to compute the scattering intensity and consequently the scattering coefficient. The radiative transfer formulation takes into account the irregular ground surface by including the rough soil surface in the boundary condition. Thus, the scattering model accounts for volume scattering inside the vegetation layer, the surface scattering from the ground and the interaction between scattering from the soil surface and the vegetation volume. The contribution to backscattering by each of the three scattering mechanisms is illustrated along with the effects of each layer or surface parameter. The major difference between the two types of vegetation is that when the incident wavelength is comparable to the size of the leaf there is a peak appearing in the mid angular region of the backscattering curve for the disc type leaf whereas it is a dip in the same region for a needle type leaf.

Fung, A. K.↗

NASA NTRS1985

Inverse methods in rough-surface scattering

For the case of unknown surfaces, which can only be studied by means of remote sensors, roughness scale relative to incident wavelength can be determined through the examination of angular scattering characteristics in both vertically and horizontally polarized states. Generally, an angular backscattering curve without polarization dependence indicates scattering by roughness scales that are larger than the incident wavelength. Over angular regions where both vertical and horizontal returns are well separated, the scattering must be dominated by roughness scales that are small by comparison with wavelength.

Fung, A. K.↗

NASA NTRS1985

The effect of Fresnel phase on scattering from a vegetation layer

Models for scattering from a vegetation layer treated as a collection of discrete scatterers usually assume far field interaction among scatterers. In a real vegetation medium such as a deciduous forest it is not always true that each leaf is in the far field of the others. This paper examines the additional effect when scatterers are permitted to be in the Fresnel zone. It is found that this in general causes the backscattering coefficient to be higher and also to have a larger rate of increase with frequency than the case of purely far field interaction.

Lee, K. K.↗

NASA NTRS1985

A multiple scattering theory for EM wave propagation in a dense random medium

For a dense medium of randomly distributed scatterers an integral formulation for the total coherent field has been developed. This formulation accounts for the multiple scattering of electromagnetic waves including both the twoand three-particle terms. It is shown that under the Markovian assumption the total coherent field and the effective field have the same effective wave number. As an illustration of this theory, the effective wave number and the extinction coefficient are derived in terms of the polarizability tensor and the pair distribution function for randomly distributed small spherical scatterers. It is found that the contribution of the three-particle term increases with the particle size, the volume fraction, the frequency and the permittivity of the particle. This increase is more significant with frequency and particle size than with other parameters.

Karam, M. A.↗

NASA NTRS1985

Propagation of waves in a bounded random layer with laminar structure

A closed form solution has been developed to obtain the intensity propagating in a bounded layer with laminar structure. Then, the brightness temperature due to an arbitrary temperature profile has been derived. Results are specialized to a half space to compare with those reported in the literature.

Karam, M. A.↗

NASA NTRS1984

Scattering models in the microwave regime

Results of first and second year research efforts are summarized in a series of complete articles and abstracts. The goal of the first year efforts was to calculate scattering from an inhomogeneous layer with irregular boundaries to model natural terrains. The model was applied to interpret measurements from vegetation, snow, and sea ice. The goal of the second year was to extend the scattering model to handle disc shaped scatterers which are comparable to incident wavelength and to use the model to investigate the relative merits between active versus passive sensing of soil moisture over vegetated terrain.

Fung, A. K.↗

NASA NTRS1984

Scattering from randomly oriented scatterers of arbitrary shape in the low-frequency limit with application to vegetation

A general theory of intensity scattering from small particles of arbitrary shape was developed based on the radiative transfer theory. Upon permitting the particles to orient in accordance with any prescribed distribution, scattering models can be derived. By making an appropriate choice of the particle size, the scattering model may be used to estimate scattering from media such as snow, vegetation and sea ice. For the purpose of illustration only comparisons with measurements from a vegetated medium are shown. The difference in scattering between elliptic and circular shaped leaves is demonstrated. In the low frequency limit, the major factors on backscattering from vegetation are found to be the depth of the vegetation layer and the orientation distribution of the leaves. The shape of the leaf is of secondary importance.

Karam, M. A.↗

NASA NTRS1984

Scattering from randomly oriented circular discs with application to vegetation

A vegetation layer is modeled by a collection of randomly oriented circular discs over a half space. The backscattering coefficient from such a half space is computed using the radiative transfer theory. It is shown that significantly different results are obtained from this theory as compared with some earlier investigations using the same modeling approach but with restricted disc orientations. In particular, the backscattered cross polarized returns cannot have a fast increasing angular trend which is inconsistent with measurements. By setting the appropriate angle of orientation to zero the theory reduces to previously published results. Comparisons are shown with measurements taken from milo, corn and wheat and good agreements are obtained for both polarized and cross polarized returns.

Karam, M. A.↗

NASA NTRS1984

A comparison between active and passive sensing of soil moisture from vegetated terrains

A comparison between active and passive sensing of soil moisture over vegetated areas is studied via scattering models. In active sensing three contributing terms to radar backscattering can be identified: (1) the ground surface scatter term; (2) the volume scatter term representing scattering from the vegetation layer; and (3) the surface volume scatter term accounting for scattering from both surface and volume. In emission three sources of contribution can also be identified: (1) surface emission; (2) upward volume emission from the vegetation layer; and (3) downward volume emission scattered upward by the ground surface. As ground moisture increases, terms (1) and (3) increase due to increase in permittivity in the active case. However, in passive sensing, term (1) decreases but term (3) increases for the same reason. This self conpensating effect produces a loss in sensitivity to change in ground moisture. Furthermore, emission from vegetation may be larger than that from the ground. Hence, the presence of vegetation layer causes a much greater loss of sensitivity to passive than active sensing of soil moisture.

Fung, A. K.↗

NASA NTRS1984

Evaluation of the radar response to land surfaces and volumes: Examination of theoretical models, target statistics, and applications

Four areas of L-band radar remote sensing of terrain were examined: (1) the behavior of the radar backscatter coefficient of distributed surface and volumes as a function of the targets' dielectric and geometric parameters and as a fnction of their physical parameters; (2) the correspondence of the angular behavior of the relative backscatter coefficient as extracted from SIR-B digital imagery and truck mounted L-band scatterometer measurements for about 100 fields; (3) the statistical behavior of SIR-B image density for targets that appear homogeneous on Thematic Mapper (TM) optical imagery and/or color IR photography; and (4) the applicability of SIR-B imagery both alone and in conjunction with TM imagery for the classification and monitoring of land cover and renewable resources.

Ulaby, F. T.↗

NASA NTRS1984

A scatter model for vegetation up to Ku-band

A scatter model is developed based on the matrix doubling method for volume scattering and the Kirchhoff method in rough surface scattering. Scattering from vegetation is assumed to be dominated by leaves and a single leaf is modeled by a thin dielectric disk. In developing the phase matrix for the disk, field within the disk is taken to be constant over the disk thickness, but phase changes across the surface of the disk are accounted for. Comparisons of this scatter model with radar measurements indicate good agreements in polarization, angular trends, and frequency up to Ku-band. This represents a considerable improvement over low-frequency scatter models which are valid up to S-band.

Eom, H. J.↗