NAIC -- NRAO School on Single - Dish Radio Astronomy: Techniques and Applications

School Abstracts:


Basic Principles of Single-Dish Radio Telescopes

Paul F. Goldsmith (NAIC)

This lecture covers the basics of how a radio telescope collects radiation from the "radio sky". We begin with a definition of key radio astronomical terminology, including the brightness distribution, brightness temperature, point-like and extended sources, flux density, and antenna temperature. In order to analyze radio telescope performance, we introduce the feed system which couples the receiver to the antenna. We continue by developing the concepts of the antenna as a phase transformer and the aperture plane field distribution, invoking the reciprocity principle to replace the receiver with a transmitter and treating the radiation pattern that would be produced by the resulting system. Using the distribution of the electric field in the aperture plane, we calculate the sensitivity of the radio telescope to a point source. This discussion is based on the antenna aperture efficiency and includes its dependence on the characteristics of the feed system, on the blockage, on systematic errors due to feed or secondary reflector defocus, and on random errors. The aperture plane field distribution is the starting point for using the Fourier Transform relationship to compute the far-field response, or power pattern, of the antenna. We analyze the effects of the telescope characteristics on the beam width and sidelobe level of the antenna's power pattern, and on the coupling efficiency to an extended source. Understanding this basic behavior allows us to determine what measurements are needed to calibrate the antenna, and to determine whether the system is performing properly.

The Measurement Process with Fully-Filled Apertures

Donald B. Campbell (NAIC)

This lecture will concentrate on how a single dish is used to obtain information about the sky brightness distribution including point and extended sources. There will be discussions of: 1) the limitations on the information that can be obtained due to the angular size of the antenna far field response or power pattern; the convolution of the sky brightness distribution with the power pattern and the concept of spatial Fourier components of the sky brightness distribution and antennas as filters of these Fourier components; 2) sources of noise, the concept of system temperature and the detectability of weak signals in the presence of noise; 3) the detectability of weak sources in the presence of "confusion" from other sources; 4) simple receiver systems, detectors and recording techniques; 4) digitization of analogue signals and the Nyquist sampling rate.

Why Single-Dishes?

Darrel Emerson (NRAO)

o Single dishes (SDs) give full spatial frequency coverage up to D/lambda. Large area surveys of very broadly distributed celestial components, e.g. Galactic HI, background continuum, etc.

o The different spatial frequency responses of SDs and interferometers, (especially in the overlap region). Picking "the right instrument for the right job".

o SDs can provide the data to enhance short spatial frequency coverage for wide-field synthesis imaging. ALMA, the first instrument designed to make single-dish measurements as well as interferometry.

o The ability to use a large collecting area on transient phenomena (e.g. pulsar single pulses, solar bursts, flare stars, SETI, etc.)

o For any frequency band, an SD needs only a single receiver; as this is a single device it can be made the very best possible. SDs can achieve the maximum radio frequency coverage with a small suite of appropriate receivers, limited only at the high-frequency end by the dish rms.

o For spectral-line observing, SDs usually have only a single data stream to process, so very large numbers of frequency channels (i.e. high frequency resolution) can be simply and economically applied on that data stream.

o SDs in finder surveys for high-resolution (interferometry) follow ups (e.g. HI in galaxies, megamasers, OH/IR star catalogs).

o Flexibility of SDs means that it is easy to make real time decisions while observing.

o Big SDs can be co-opted on occasion into high sensitivity VLBI arrays, (i.e. EVN & Global Arrays, ad-hoc arrays.)

o SDs and planetary radar.

The Receiver System - cm Regime

Roger D. Norrod (NRAO)

The receiver front-end of a radio telescope is generally considered to encompass components which amplify, filter, and frequency convert signals provided by the antenna to a level and frequency range appropriate for detection. This presentation will discuss critical parts of the centimeter wave radio astronomy front-end and factors impacting the design and performance. The feed efficiently converts propagating electromagnetic fields near a reflector antenna's focal point to a guided wave in coax or waveguide. Some types of feeds inherently detect and separate polarizations; other types require an orthomode transducer to deliver orthogonal polarizations to separate channels. Low-noise amplifiers, usually cryogenically cooled, amplify the signal and set the receiver noise level, and are followed by filters, mixers, and additional amplification. All the passive and active components add electrical noise to the signal, and models used during receiver design will be presented, explaining why loss and noise introduced in the early stages of the receiver are critical. The linear operating range of active components is limited by their power handling capacity, and how these limitations are considered will be discussed. We will also discuss stability of the receiver, and practical means to achieve the required performance. Finally, the presentation will briefly cover related subsystems such as local oscillators and cryogenics.

The Receiver System - mm Regime

John Payne (NRAO)

The millimeter wavelength radio astronomy band is now generally taken to include frequencies of approximately 60 GHz to around 300 GHz, and the so-called sub-mm band extends this up to frequencies of around 1000 GHz (1 THz). Unlike centimeter-wave, ground-based radio astronomy, the mm/sub-mm frequency range is limited by the properties of the atmosphere which are briefly described.

The difficulties of constructing highly sensitive receivers for these high frequencies are described along with the commonly adopted solutions. Particular emphasis is given to the modern receivers such as those being planned for the ALMA interferometric array destined for installation on a 5000-meter-altitude site in the Atacama desert in Chile. These receivers use super conducting devices which require temperatures of around 4k to operate satisfactorily. Brief descriptions of the various sub-systems needed to construct such a receiver will be given.


Backends (part 1)

Rick Fisher (NRAO)

The final stages of receiver electronics are designed to extract information about the intensity of cosmic signals as a function of time, frequency, and polarization. The required signal processing can be as simple as a total power detector or as complex as a pulsar search machine that looks for periodic, dispersed pulse signatures in the time and frequency domain. This mini-lecture will give a brief overview of square-law detectors, FFT spectrometers, polarimeters, and pulsar processors. I will touch on calibration issues and the synchronous control of front-end calibration signals and beam and load switches.

Noise Statistics and Spectral Analysis

Jon Hagen (NAIC)

The output of a radio astronomy receiver is a time varying, random valued voltage with a mean value of zero. This voltage can be partially described by its power, the mean square value. The voltage can be completely described by its autocorrelation function, R(tau), defined as the average value of the product of pairs of voltage samples taken at times separated by tau seconds. Note that the power is just R(0). Special purpose back-end processors form the averages to estimate hundreds or thousands of points on the autocorrelation function (ACF). This estimate of the ACF is then Fourier transformed to produce an estimate of the power spectral density function (PSD) wherein emission from a molecular line will show up as an obvious peak. Alternatively, the PSD can be estimated directly by Fourier transforming successive blocks of voltage samples and then averaging the magnitudes of the respective Fourier components.

Because radio astronomy signals have Gaussian amplitude statistics, it turns out that the ACF can be accurately estimated even when the signal has been sampled by a digitizer of very low precision. The extreme case is 1-bit sampling, where only the polarity of the voltage is measured. The advantage of coarse quantization is hardware simplicity. The cost is an increase in averaging time. The procedures used to convert the ACF of a coarsely sampled voltage into the ACF of the undistorted original voltage are generally called "Van Vleck" corrections, after J.H. Van Vleck, who was the first to analyze the 1-bit case.


Stray Radiation and How to Deal with It

Felix J. Lockman (NRAO, Green Bank)

An antenna has response in all directions, not just the direction that it is ``pointing''. When using an antenna for radio astronomy, some signals which enter the receiver and are detected are ``stray'' in that they come through a sidelobe and not the main beam. Terrestrial interference is one example, but the term ``stray'' is usually used for signals of natural origin which might be confused with the object under study. A large fraction of the stray radiation can be eliminated in the standard observational techniques used to calibrate continuum or line intensities. In some experiments, however, the stray signals constitute the fundamental limitation on the accuracy of the measurements, and can result in errors of more than 100%.

Every single-dish measurement suffers to some extent from stray radiation, so it is important to recognize the phenomenon and understand the circumstances in which it becomes important. This talk will describe the general phenomenon of stray radiation, illustrate some specific circumstances where it must be considered, and consider some of the techniques, often rather onerous, that have be used to reduce its effects.

Continuum 1: General Aspects

Jim Condon (NRAO)

(My intention is to cover the basic principles of continuum observing, present simple equations, data, and references to help the novice continuum observer plan and reduce single-dish observations, and reference specialized papers for those wishing to pursue a topic further.)

o What is a continuum source?

o Why observe continuum sources?
-- To measure source flux density, position, etc.
-- As calibrators for other types of observations.

o How to measure surface brightness and flux density.

o How to measure source positions.

o How to measure polarization (very brief, since this will have been covered by C. Heiles).

o Discrete source surveys.

o Why use a single dish instead of an interferometer?

o Ideal radiometer equation.

o Nonideal effects:
-- receiver gain fluctuations
-- atmospheric emission
-- ground radiation
-- scintillation (ionosphere, IPM, ISM)
-- confusion by discrete sources

Continuum 2: Specific Applications

Chris Salter (NAIC)

(This talk will compliment "Continuum 1" by covering more specific observing/analysis techniques.)

o Effects of RFI (broadband impulsive and narrowband CW), use of spectrometer to remove narrowband RFI.

o Effects of atmosphere (why does continuum need to position switch much more rapidly than spectral-line observing)

o Total-power receiver (short-comings), Dicke switching (e.g. load switching, sky-horn switching, double Dicke switching, Ryle-Vonberg system, nutating subreflectors, wobblers, etc.), correlation receiver, software beam switching.

o Mapping: Scanning, basket weaving, Emerson, Klein, & Haslam algorithm and other developments. Array detectors.

o Comparison of continuum images (spectral index distributions, rotation measure distributions).

Spectral-Line 1: General Aspects

Harvey S. Liszt (NRAO)

The use of spectral lines in radio astronomy was conceived by Oort as a means of mapping the large-scale structure of the Milky Way, free from the obscuration of interstellar dust and earthly weather. Now, as we celebrate the 50th anniversary of the discovery of the H Iline, spectral line radio astronomy has detected thousands of different transitions from hundreds of atomic and molecular species in interstellar and circumstellar gas, neutral and ionized, near and far.

The line profiles we take home from the radiotelescope originate when local processes on the scale of individual atoms cause the emission and absorption of photons. These photons propagate through the medium and do not emerge (if at all) until they have run quite a gauntlet. There is absolutely no guarantee that the emergent spectrum represents conditions deep inside the gas. Worse, the viewing geometry and our sensing of only one component of the gas motion (projected onto the line of sight) can cause signals from various parts of the medium to blend together at the same received frequency (the opposite can also occur).

Of course the good news about these sensitivities, which accrue on a range of scales differing by 30 orders of magnitude (from nm to kpc) is that we can hope to measure an astonishingly wide range of properties, if only we can interpret our data. In the hope of providing even the barest introduction to this field, I'll pick a few representative combinations of source medium, line species, viewing geometry, earthly observing conditions and receiving equipment, and analyze them at varying levels of depth and sophistication.


Advanced Spectral Line Topics

John Dickey (University of Minnesota)

We will discuss spectral line mapping, and especially the interpretation of maps and spectral cubes. I will concentrate on three areas,
-- observational techniques for making spectral line maps,
-- the spectral line cube and its moments,
-- interpretation of patterns in the cube, especially for rotating disks.
By the end of the class, I hope students will be ready to make and interpret maps of the HI emission from nearby galaxies, HVC's, or Galactic ISM structures for their practicum.

Observational techniques for spectral line mapping include the simple questions of sensitivity, bandwidth, resolution, and integration time, plus more problematic questions such as bandpass calibration, continuum subtraction, gridding, and absolute brightness and flux calibration. I will show some examples of how we calibrate "on-the-fly" mapping data from the Parkes and Arecibo telescopes. Briefly I will discuss combining maps from interferometer and single dish telescopes.

Understanding spectral line cubes is the main goal of my class. I will illustrate structures in cubes with several real data sets of both Galactic and extra-galactic HI maps. I hope to use the KARMA package for this, with the students working directly with the data in small groups, to do some quick exercises. Then I will formally define the moments of the cube and illustrate them with data. We will discuss the best ways to compute moments (robust against noise and interference) and what the moments mean physically.

Finally as an important example of using moment maps, we will consider the case of a gaseous disk in differential rotation, as in a spiral galaxy. We will discuss the shapes of rotation curves, the significance of the velocity dispersion, and the column density distribution as measured with the 21-cm line for spiral galaxies. We will also consider how best to fit the several parameters of a model for a rotating disk, and the characteristic signatures of errors in the parameters which appear in the residual map.


Pulsar Observations: Propagation Effects, Searching, Distance Estimates, Scintillations and VLBI

Jim Cordes (Cornell University)

The role of propagation effects (dispersive propagation, scattering, Faraday rotation and HI absorption) will be discussed, particularly as they affect sensitivities of pulsar searches. We will summarize dedispersion techniques that manipulate "postdetection" signals from analog and digital filter banks. Search data-acquisition and algorithms for detecting isolated and binary pulsars will be discussed. Interstellar scintillation, caused by multipath scattering in the interstellar medium (ISM), influences search sensitivities but also provides unique information about turbulence in the ISM. Data acquisition and analysis of scintillations will be summarized. Dispersion and scattering provide important input to models for the electron density in the Galaxy. The most recent electron density model will be presented. Finally, the scientific importance of very-long-baseline-interferometry of pulsars, along with pulsar-specific VLBI techniques, will be outlined.

Pulsar Observations: Coherent Dedispersion, Polarization, and Timing

Ingrid Stairs (NRAO)

Pre-detection, or "coherent," dedispersion is a powerful technique which completely eliminates the effects of dispersive smearing on pulsar profiles, greatly increasing the precision of timing and polarization observations. Various software and hardware implementations of this method are described. Polarization observations with filterbanks and with coherent dedispersion are discussed, including calibration and compensation for instrumental effects. Finally, the technique of pulsar timing is presented, with descriptions of the method for obtaining pulse times of arrival and the software used to fit models of the pulsar's spin-down behavior, and a broad overview of scientific goals.

Planetary Radar

Greg Black (NRAO)

Radar is a powerful tool for studying the Solar System, with its reach limited only by the transmitter power available. It has been used to observe targets ranging in size from the rings of Saturn down to house-sized asteroids. Since an observer has control of the illumination source, a radar experiment provides information not available from passive observing methods. On centimeter to meter scales it is a sensitive probe of surface characteristics such as dielectric constant and roughness, and on larger scales can map topography and determine shapes of irregular objects at resolutions finer than other ground-based methods. This lecture will cover the basic techniques of planetary radar astronomy, give an overview of the scientific questions that can be addressed, and survey some recent results. Key points of the lecture will be: the radar equation; principles and benefits of modulating the transmitted signal; data processing; and an outline of current radar systems' parameters.

Fundamentals of Polarization Measurement

Carl Heiles (University of California Berkeley)

The polarization of electromagnetic radiation, no matter what wavelength, is best specified by Stokes parameters because they are independent, can be arithmetically manipulated, and the noise statistics are Gaussian. However, as happens in all kinds of measurements, there are instrumental effects that must be accounted for.

Instrumental effects in polarization measurement, no matter what wavelength, break down into several types. One is the relationship between the measured Stokes parameters and the true ones; this is best specified by the 4 X 4 element Mueller matrix, which describes the transfer function of the instrumentation. Another is the stability of these matrix parameters, which translates directly into the accuracy of polarization measurement. Another is the variation of the matrix parameters within the telescope beam or field of view, which affects directly the accuracy with which polarization of extended sources having angular structure can be measured.

We will discuss the fundamentals behind these issues using simple examples and phenomenological illustrations, with a minimum of mathematical emphasis. Radio astronomy is unique in allowing all four Stokes parameters to be measured simultaneously with no loss in signal/noise by using cross correlation, if the relevant instrumentation is available. We will describe the practical issues involved and offer recommended techniques for specific cases of polarization measurement. Perhaps the most important application is accurate measurement of the total intensity (the Stokes I parameter), which is the sum of two orthogonal polarizations, and we will offer our perspective on why current commonly-used techniques are not always optimum.


Calibration techniques at radio wavelengths

Karen O'Neil (NAIC)

Why calibration is needed

Atmospheric Opacity (briefly, as it will be covered later)

Calibration scale definitions
o System temperature
- How to measure
o Flux Density conversion
- Rayleigh-Jeans approximation

Antenna efficiency measurements
o Aperture efficiency
- definition, measurement techniques
o Beam efficiency
o Coma/sidelobes
o Pointing/Focus Issues

Temperature scale calibration techniques
o Pulsed noise cal
o Load cal

Baseline/Flux calibration
o Position/Frequency Switching
o Flux density standards
o Gain curves
o Standard sources - line/continuum

Error Budget

Reference Papers & Resources
o Kraus
o Baars paper
o etc.


Calibration Techniques at Millimeter Wavelengths

P. R. Jewell (NRAO)

Calibration practices for millimeter wavelengths are somewhat different than that for meter and centimeter wavelengths. There are technological, atmospheric, and historical reasons for this. This lecture will review the specific techniques used for millimeter wave calibration and will highlight the differences between these and the techniques used at longer wavelengths. Cases for which calibration techniques used at different wavelengths might be merged or rationalized will be discussed. The importance of the atmosphere at centimeter, millimeter, and submillimeter wavelengths will be discussed in detail. Topics will include specific calibration techniques such as the hot/sky chopper wheel method, variations such as hot/cold/sky schemes, sky tipping calibration, and the possibilities of subreflector based calibration sources. Calibration loss factors including rear and forward spillover and error beam losses will be described and illustrated. The TA*, TR*, and TMB temperature scales will be defined. Techniques for absolute calibration, and the effects of double sideband versus single sideband operation will also be discussed.

Reduction and Analysis Techniques

R. Maddalena (NRAO)

Single-dish observations can be made in a myriad of ways with different observing technique almost always requiring different kinds of data analysis. We will cover in this class the standard and basic continuum and spectral line analysis algorithms common to all single-dish data analysis packages. However, we will not cover the very specialized fields of polarimetry, pulsar, or radar data reduction. In the case of continuum observations the student will learn the steps used to derive the flux of a point source as well as the more common data analysis techniques for generating and analyzing maps of extended sources. For spectral line data, we will discuss how the analysis of an observation will depend upon the backend type (filter-bank, autocorrelation, or AOS) and observing technique (frequency-, position-, or beam-switched). We will concentrate on the analysis algorithms usually applied to single spectra (bandpass and velocity calibration, data averaging and smoothing, baseline fitting, component fitting, ...) and how to produce and analyze spectral-line data cubes.

AIPS++ DISH tutorial

Joe McMullin (NRAO)

DISH is a package within AIPS++ that provides a streamlined environment for working with single dish radio astronomy data. The priority in the development of DISH has been to produce traditional spectral line analysis tools with the flexibility and extensibility that is a key attribute of the AIPS++ package.

The tutorial will provide a census of the components of the package (GUI, plotter, logs, history, etc), an overview of the available reduction tools (flagging, averaging, regridding, imaging, etc), and worked examples of different experiments from start to finish. The interface, both GUI and command line, will be discussed in detail, including ways to expand the tool-kit through the use of Glish scripts.

Documentation is available via: http://www.nrao.edu/daily/docs/aips++.html CDs with the software will also be provided.


The Effects of the Atmosphere and Ionosphere

Luca Olmi (University of Massachusetts)

In this lecture I will mainly discuss the effects of the neutral atmosphere, and to a lesser extent of the ionosphere, on radio astronomical observations carried out with single-dish telescopes from the surface of the Earth. We are concerned with three types of effects: large-scale refractive effects, absorption, and scattering by the turbulent structure in the media. The phenomenon of scattering results in "seeing", and creates refractive index variations which limit the resolution and sensitivity of observations of astronomical sources. In the troposphere, water vapor plays a fundamental role in radio propagation, the refractivity of water vapor being about 20 times greater in the radio range than in the near-infrared or optical regimes. As a consequence, phase fluctuations at frequencies higher than about 1 GHz are predominantly caused by fluctuations in the distribution of water vapor, and I will thus concentrate this lecture on tropospheric-induced radio seeing. On filled-aperture telescopes, radio seeing shows up as an anomalous refraction (AR), i.e. an apparent displacement of a radio source from its true position. The magnitude of this effect, as a fraction of the beam width, is bigger on larger telescopes, and thus its impact on pointing is likely to become critically important in the next generation of electrically large, filled-aperture radio telescopes. I will thus present the results of recent, systematic AR measurements and discuss a model study of AR effects, obtained via numerical simulations of two-dimensional phase screens. I will finally discuss the basic concept and requirements of a tip-tilt compensation system at millimeter wavelengths, and will also present a proposed design based on a scanning microwave radiometer as a wave front sensing device.

RFI and How to Deal with It

Rick Fisher (NRAO)

Astronomers share the radio spectrum with a multitude of other users who transmit useful signals with a wide variety of spectral and temporal characteristics. Even portions of the spectrum allocated for exclusive use by radio astronomy are subject to contamination by incidental radiators such as computers, digital cameras, and observatory test equipment. Much of the spectrum that is not allocated for radio astronomy is also available to us, but it can take some careful planning to obtain useful data. Important observing parameters include time of day, receiver and spectrometer dynamic range, and temporal resolution. This lecture will discuss a number of software tools have been and are being developed to help recognize and remove interference from astronomical data. I will also mention a few signal processing techniques that are being developed to remove interference coherently or with time resolutions much greater than can be realized in software.

Spectrum Management

Tapasi Ghosh (NAIC)

Life without the use of radio frequencies is unimaginable today. The frequency range involved, and the mode of its usage, is diverse and ever-changing with the developments of the latest technology. In the midst of all this, there is a minority of spectrum users who are labeled "passive". Radio astronomers and remote sensing groups come under this heading, having no control over the signals that they attempt to receive. It is of the utmost importance that all spectrum users may operate and evolve in a manner of peaceful coexistence. This is the goal of spectrum management.

As radio waves "do not know" any national boundaries, such management issues have to be agreed upon globally. The mechanisms that have been set up for this purpose, with their global and national counterparts, will be described in this talk. How, as a minority, the interests of radio astronomers may be best served, both now and in future, will also be discussed.


Short-Spacings Correction from the Single-Dish Point of View

Snezana Stanimirovic (NAIC)

While, in general, interferometers provide high spatial resolution for imaging small-scale structure (corresponding to high spatial frequencies in the Fourier plane), single-dishes can be used to image the largest spatial scales (corresponding to the lowest spatial frequencies), including the total power (corresponding to zero spatial frequency). For many astrophysical studies, it is essential to bring `both worlds' together by combining information over a wide range of spatial frequencies. This lecture will demonstrate the effects of missing short-spacings, and concentrate on two main issues: (a) how to provide missing short-spacings to interferometric data, and (b) how to combine short-spacing single-dish data with that from an interferometer.

Focal Plane Arrays

John Payne (NRAO)

Over the past few years, the sensitivity of mm/sub-mm receivers has improved dramatically. We are now at the point of the atmosphere and other inevitable sources of noise being equal to, or perhaps greater than, the noise contributed by the receiver itself. Under such conditions, the observing efficiency (for mapping extended regions) may be improved by adding receivers in the focal plane in the manner of the CCDs used in optical astronomy. Details of the principles involved and brief descriptions of existing systems will be given.

Bolometers for Millimeter-Wave Astronomy

Wayne S. Holland & William D. Duncan (ATC)

During the past decade considerable progress has been made in the development of bolometric detectors for submillimetre and millimetre-wave astronomy. With the introduction of imaging arrays these wavebands have been undergoing a major revolution with many new and exciting discoveries being made in almost all disciplines of astronomy.

This lecture will focus on the need for deep, wide-field continuum imaging in the submm/mm regime. The basic principles of bolometry will be explained, together with the "figures of merit" that characterise bolometer performance. Particular examples will focus on the problems of using wide-bandwidth devices in conditions where the atmospheric background can change on short timescales.

Practical devices - from single pixels to the current imaging arrays - will be discussed, with particular emphasis on the need for ultra-low temperature operation for sky-background limited performance. Examples will be given of how current instrument perform on ground-based telescopes, including the issue of optimum coupling of the small detector element to a large millimetre-wave telescope.

The development of integrated filled arrays of many thousand pixel cameras bodes well for the future. The new generation of both ground-based and space-borne instruments will also be described, as well as other exciting prospects for the future.

 


Last modified: 3-April-2001
Questions, comments...