Interferometry and separated collecting elements

An array compares radiation received at different locations

Interferometry combines radiation information from separated collecting elements. A wave arriving from a particular direction reaches their locations with a direction-dependent relative path and phase. Measuring calibrated correlations can therefore constrain how emission is distributed over angle.

Phase describes position within a wave cycle. A baseline is the vector separation between elements; its projection relative to the source direction matters. The observing system must account for the relevant delays, instrumental response and phase behaviour rather than merely place two independent pictures side by side.

For a suitably calibrated, limited-field model at a selected wavelength, correlations sample angular spatial-frequency information. A spatial frequency describes how rapidly a pattern changes across angle. A single baseline supplies a limited pattern constraint, not a complete two-dimensional sky image.

Different projected baseline directions and lengths provide complementary information. Earth’s rotation can change those projections during an observation. Turning the measurements into an image requires a response and reconstruction model with uncertainty and incomplete coverage retained.

Wavelength divided by projected baseline gives a useful scale

Let λ be observing wavelength and B the relevant projected baseline length, in the same length unit. The characteristic angular fringe scale λ/B is in radians. It describes variation along the corresponding projected direction under the stated geometry.

This is not automatically the full width of a reconstructed image response or an exact universal Rayleigh coefficient. The collection of baselines, their weights, calibration and imaging method determine the practical response and distinguishable structure.

In an original synthetic layout, two ideal clear circular 10 m elements observe at λ =3 mm =0.003 m. Compare transverse projected baselines 100 m and 200 m, with the source direction perpendicular to the plan-view drawing.

For 100 m, λ/B =3 ×10⁻⁵ rad, about 6.19 arcseconds. For 200 m it is 1.5 ×10⁻⁵ rad, about 3.09 arcseconds. Doubling this baseline halves the characteristic scale while leaving the two physical collecting elements unchanged.

The elements and their separation have different roles

Each 10 m ideal circular element has area π ×5² m². Their combined clear geometric area is 50π m², about 157.08 m², in both layouts. The space between them is not a filled collecting surface.

Moving the same elements farther apart can improve access to finer angular spatial structure in the projected direction without increasing that geometric area. It does not make the array equally sensitive to a filled dish of diameter equal to the baseline.

Each individual antenna also has a directional primary response that helps set the usable field of one pointing. The element diameter and wavelength contribute to that response; it is distinct from the fine characteristic scale associated with a long baseline.

The figure therefore draws the same element sizes on one common metres scale in both rows, with centre-to-centre baseline bars. It supplies no invented observed sky image or synthesised beam. The angular values are calculated characteristic scales, not a claim of a detected or resolved real source.

An image needs more than one long separation

With N distinct antennas there are N(N −1)/2 unordered pairs. Two antennas give one pair, three give three, and four give six. This is a combinatorial count, not a count of independent image pixels or a guarantee of complete angular coverage.

Baselines can be redundant or sample only limited directions and lengths. Missing short spacings can leave weak constraints on broad extended emission, even when long baselines provide fine angular information. Compact configurations and complementary total-power measurements can supply information of a different scale.

Calibration must also address receiver behaviour, relative timing, phase changes, atmosphere and averaging. A longer baseline is not useful merely because its geometric number is large; the measured correlations must preserve the information the model assumes.

ALMA’s imaging explanation illustrates the role of combining configurations and measurement types in an international observing system. The synthetic two-element layout here remains a separate teaching model, with no current facility specification or performance ranking implied.

The next original chapter examines observatories, surveys and archives. Those observing programmes depend on the instrument responses, sampling and calibration developed here; their catalogues retain the limits of how the evidence was collected.

Worked examples

Try the reasoning

1. How many distinct unordered antenna pairs are available from four antennas?

   Show the answer and reasoning

   Six.

   4 ×3/2 =6. Coverage, redundancy and calibration still matter.

2. Does a 200 m baseline between two 10 m elements create a filled 200 m collecting dish?

   Show the answer and reasoning

   No.

   The elements collect radiation; the separation contributes relative spatial information.

3. Can an array with fine long-baseline information still have limited constraints on extended emission?

   Show the answer and reasoning

   Yes.

   Missing short spacings can leave large angular structures poorly measured.

4. Does λ/B alone define field of view, total sensitivity and a complete image?

   Show the answer and reasoning

   No.

   Primary response, collecting elements, sampling coverage, calibration and reconstruction supply different parts of the measurement.

What to carry forward

• Separated elements add direction-dependent relative information through calibrated correlations.
• Baseline angular scale, collecting area and primary field are distinct quantities.
• An image needs appropriate baseline coverage, calibration and reconstruction, including attention to missing angular scales.