Radio Sources and Scintillation

A review is given of the interplay between studies of compact radio sources and the scattering and scintillations that occur as the signals travel through the irregular refractive index of the interstellar and interplanetary plasmas. This revised version was published online in July 2006 with corrections to the Cover Date.     

Disturbances in the solar wind from IPS measurements in August 1972

From IPS and spacecraft measurements of the solar wind combined with geomagnetic observations, we identify the passage of three main disturbances through the solar wind from solar flares on August 2, 4 and 7. From a detailed study of the IPS data covering the third event, we conclude that the extent of the disturbance front at 1 AU covered about ±60° in longitude and more than 30° in latitude from the flare normal. If interpreted as a blast wave according to the model of De Young and Hundhausen (1971), the disturbance was ejected from the Sun into a cone of half-angle 45°±15°.     

The inner scale of the plasma turbulence towards PSR J1644−4559

By measuring the decaying shape of the scatter-broadened pulse from the bright distant pulsar PSR J1644−4559, we probe waves scattered at relatively high angles by very small spatial scales in the interstellar plasma, which allows us to test for a wavenumber cutoff in the plasma density spectrum. Under the hypothesis that the density spectrum is due to plasma turbulence, we can thus investigate the (inner) scale at which the turbulence is dissipated. We report observations carried out with the Parkes radio telescope at 660 MHz from which we find strong evidence for an inner scale in the range 70–100 km, assuming an isotropic Kolmogorov spectrum. By identifying the inner scale with the ion inertial scale, we can also estimate the mean electron density of the scattering region to be 5–10 cm⁻³. This is comparable with the electron density of H  ii region G339.1−0.4, which lies in front of the pulsar, and so confirms that this region dominates the scattering. We conclude that the plasma inside the region is characterized by fully developed turbulence with an outer scale in the range 1–20 pc and an inner scale of 70–100 km. The shape of the rising edge of the pulse constrains the distribution of the strongly scattering plasma to be spread over about 20 per cent of the 4.6 kpc path from the pulsar, but with similarly high electron densities in two or more thin layers, their thicknesses can only be 10–20 pc.     

A model‐based data‐driven dictionary learning for seismic data representation

Planar waves events recorded in a seismic array can be represented as lines in the Fourier domain. However, in the real world, seismic events usually have curvature or amplitude variability, which means that their Fourier transforms are no longer strictly linear but rather occupy conic regions of the Fourier domain that are narrow at low frequencies but broaden at high frequencies where the effect of curvature becomes more pronounced. One can consider these regions as localised “signal cones”. In this work, we consider a space–time variable signal cone to model the seismic data. The variability of the signal cone is obtained through scaling, slanting, and translation of the kernel for cone‐limited (C‐limited) functions (functions whose Fourier transform lives within a cone) or C‐Gaussian function (a multivariate function whose Fourier transform decays exponentially with respect to slowness and frequency), which constitutes our dictionary. We find a discrete number of scaling, slanting, and translation parameters from a continuum by optimally matching the data. This is a non‐linear optimisation problem, which we address by a fixed‐point method that utilises a variable projection method with ?₁ constraints on the linear parameters and bound constraints on the non‐linear parameters. We observe that slow decay and oscillatory behaviour of the kernel for C‐limited functions constitute bottlenecks for the optimisation problem, which we partially overcome by the C‐Gaussian function. We demonstrate our method through an interpolation example. We present the interpolation result using the estimated parameters obtained from the proposed method and compare it with those obtained using sparsity‐promoting curvelet decomposition, matching pursuit Fourier interpolation, and sparsity‐promoting plane‐wave decomposition methods.     

Azimuthal variation in AVO response for fractured gas sands

Natural fractures in reservoirs play an important role in determining fluid flow during production, and hence the density and orientation of fractures is of great interest. In the presence of aligned vertical fractures, the reflection amplitude at finite offset varies with azimuth. The effect of natural fractures on the azimuthal AVO response from a gas-sandstone reservoir encased within shale is investigated. A simple expression for the difference in P-wave reflection coefficient from the top of the reservoir parallel and perpendicular to the strike of the fractures is obtained in terms of the normal and tangential compliances, Z_N and Z_T, of the fractures. This expression is valid for small anisotropy and material contrasts and is compared with the results of numerical modelling. For a given value of Z_T, the azimuthal variation in reflection coefficient at moderate offsets is found to increase with decreasing Z_N/Z_T. For gas-filled open fractures Z_N/Z_T ≈ 1, but a lower ratio of Z_N/Z_T may result from the presence of cement or clay within the fractures, or from the presence of a fluid with non-zero bulk modulus. For Z_N/Z_T = 1 and moderate offsets, the variation with offset of the reflection coefficient from the top of the fractured unit is dominated by the contrast in Poisson's ratio between the gas sand and the overlying shale, the effect of fractures only becoming noticeable as the critical angle for the unfractured sandstone is approached. However, for reflections from the base of the fractured unit, the variation in reflection amplitude with azimuth is much greater at conventional seismic offsets than for the reflection from the top. Azimuthal variations in the strength of the reflection from the top of the reservoir depend only on the variation in reflection coefficient, whereas the raypath is also a function of azimuth for reflections from the base of the fractured unit, leading to stronger, more visible, variations of AVO with azimuth. It follows that an azimuthal variation in AVO due to fractures in the overburden may be misinterpreted as due to the presence of aligned fractures in the reservoir.     

A probabilistic model for ghost delay time estimation based on recording geometry

In order to deconvolve the ghost response from marine seismic data, an estimate of the ghost operator is required. Typically, this estimate is made using a model of in‐plane propagation, i.e., the ray path at the receiver falls in the vertical plane defined by the source and receiver locations. Unfortunately, this model breaks down when the source is in a crossline position relative to the receiver spread. In this situation, in‐plane signals can only exist in a small region of the signal cone. In this paper, we use Bayes' theory to model the posterior probability distribution functions for the vertical component of the ray vector given the known source–receiver azimuth and the measured inline component of the ray vector. This provides a model for the ghost delay time based on the acquisition geometry and the dip of the wave in the plane of the streamer. The model is fairly robust with regard to the prior assumptions and controlled by a single parameter that is related to the likelihood of in‐plane propagation. The expected values of the resulting distributions are consistent with the deterministic in‐plane model when in‐plane likelihood is high but valid everywhere in the signal cone. Relaxing the in‐plane likelihood to a reasonable degree radically simplifies the shape of the expected‐value surface, lending itself for use in deghosting algorithms. The model can also be extended to other plane‐wave processing problems such as interpolation.     

Long-term scintillation observations of five pulsars at 1540 MHz

From 2001 January to 2002 June, we monitored PSRs B0329+54, B0823+26, B1929+10, B2020+28 and B2021+51 using the Nanshan 25-m radio telescope of the Urumqi Observatory to study their diffractive interstellar scintillation (DISS). The average interval between observations was about 9 d and the observation duration ranged between 2 and 6 h depending on the pulsar. Wide variations in the DISS parameters were observed over the 18-month data span. Despite this, the average scintillation velocities are in excellent agreement with the proper motion velocities. The average two-dimensional autocorrelation function for PSR B0329+54 is well described by a thin-screen Kolmogorov model, at least along the time and frequency axes. Observed modulation indices for the DISS time-scale and bandwidth and the pulsar flux density are greater than values predicted for a Kolmogorov spectrum of electron density fluctuations. Correlated variations over times that are long compared to the nominal refractive scintillation time are observed, suggesting that larger scale density fluctuations are important. For these pulsars, the scintillation bandwidth as a function of frequency has a power-law index  (∼3.6)  much less than that expected for Kolmogorov turbulence (∼4.4). Sloping fringes are commonly observed in the dynamic spectra, especially for PSR B0329+54. The detected range of fringe slopes are limited by our observing resolution. Our observations are sensitive to larger-scale fringes and hence smaller refractive angles, corresponding to the central part of the scattering disc.     

The variable projection method for waveform inversion with an unknown source function

This paper compares three alternative algorithms for simultaneously estimating a source wavelet at the same time as an earth model in full‐waveform inversion: (i) simultaneous descent, (ii) alternating descent and (iii) descent with the variable projection method. The latter is a technique for solving separable least‐squares problems that is well‐known in the applied mathematics literature. When applied to full‐waveform inversion, it involves making the source wavelet an implicit function of the earth model via a least‐squares filter‐estimation process. Since the source wavelet becomes purely a function of medium parameters, it no longer needs to be treated as a separate unknown in the inversion. Essentially, the predicted data are projected onto the measured data in a least‐squares sense at every function evaluation, making use of the fact that the filter estimation problem is trivial when compared to the full‐waveform inversion problem. Numerical tests on a simple 1D model indicate that the variable projection method gives the best result; actually producing results in quality that are very similar to control experiments with a known, correct wavelet.     

The Wilson–Burg method of spectral factorization with application to helical filtering

Spectral factorization is a computational procedure for constructing minimum-phase (stable inverse) filters required for recursive inverse filtering. We present a novel method of spectral factorization. The method iteratively constructs an approximation of the minimum-phase filter with the given autocorrelation by repeated forward and inverse filtering and rearranging of the terms. This procedure is especially efficient in the multidimensional case, where the inverse recursive filtering is enabled by the helix transform. To exemplify a practical application of the proposed method, we consider the problem of smooth two-dimensional data regularization. Splines in tension are smooth interpolation surfaces whose behaviour in unconstrained regions is controlled by the tension parameter. We show that such surfaces can be efficiently constructed with recursive filter preconditioning and we introduce a family of corresponding two-dimensional minimum-phase filters. The filters are created by spectral factorization on a helix.