Taken on January 17, 2023 from 06:14:14 to 06:21:27 UTC by the National Radio Astronomy Observatory’s (NRAO) 20-meter radio telescope in the Green Bank Observatory, WV.
About the Image
Overview
This image is an aligned projection of the supermassive black hole Virgo A in the center of the supergiant elliptical galaxy Messier 87 from the radio spectra to the visible spectrum. The returned image from the telescope was assigned arbitrary visible colors based on the frequencies detected in the radio, and the image was internally aligned down the vertical axis algorithmically to correct for the drift in the mapping pattern of the telescope between sweeps. This image is part of a larger dataset that was used to calculate the surface temperature of the moon on different dates; more information on this data can be found below and in the credits section.
Radio Ga Ga
Though radio telescopes are very big, their resolution is inherently small, so they must be moved to scan large areas of sky to create an image; for this image the telescope took 23 passes of 0.150 degrees vertically apart and 0.200 degrees each in a side-to-side map pattern. In the radio, the specific frequencies the telescope was to detect was first limited by the HI filter and further narrowed to between 1355.0 MHz and 1435.0 MHz with a center frequency of 1395.0 MHz. Additionally, in order to get a clear picture, the telescope was instructed to only begin imaging when Virgo A was twenty degrees above the horizon and ten degrees away from the sun.
Shifty Sweeps
Initially, the render of the data returned by the telescope was wonky and out of sorts: whole rows of pixels were shifted left and right of the center line. This occurs when there is a delay between when the sampled portion of the image is taken and the sample is timestamped. To correct for this, the online software used to prompt and interface with the telescope, Skynet, has a built-in radio cartography algorithm that can correct for this delay. This feature is also useful for cleaning radio images of defects and includes presets that can remove radio frequency interference (RFI) and, using a projected surface model, can remove noise in an image without making it much blurrier; this image used the “Bright Target with Airy Rings” preset and lowest weighting for the surface model so that the image was still measurement quality. Additionally, the frequency range of 1380.0 MHz to 1382.1 MHz was isolated and removed from the image entirely, because that range was contaminated with RFI. Below you can see the results of the results of the algorithm skewing the sections of the images back onto the center line. Notice above and below but particularly to the right of the object the faint blue-ish blobs in the image: theses are clouds of hydrogen that surround Virgo A and can be detected in the radio spectra.
Related Measurements
As alluded to earlier, this image was taken for a measurement as part of a larger dataset, specifically, this image was taken as a calibration to find the flux of the moon in a separate image taken in the radio spectra by the same telescope on the same night. Flux is defined as the energy output over time spread over the area being measured. Using Afterglow, a browser-based astrophotography image processing software, the center of the object in the image could be found, its area approximated, and its uncalibrated flux determined. The photometric aperture was set to 28 pixels to encompass the entirety of Virgo A in the image, and the inner and outer annuli were set to 50 and 70 pixels respectively. Afterglow determined a flux value of 1,057.745 for the object in the image, not knowing it was Virgo A, but since the goal of the researchers’ observations was to determine if the flux of the moon changes with its phases, it was important to have a calibration target that has a stable, known flux. Virgo A fits this criteria, and its actual flux is on average 229.2 Janskies (Jy), where 1 Jy is equal to 10−26 W⋅m−2⋅Hz−1. The next section discusses how these measurements are used to calibrate the image of the moon.
The Data
Context
The researchers’ objective with the data from this image was to be a calibration for data from a separate radio image of the moon taken by the same telescope on the same night. The calibrated measurement would then be compared to another calibrated measurement of the moon on a different night to determine of the flux of the moon changes with its phase. This endeavor required the four images seen below, and the data and results of these images is featured in the following sections.
at 09:01:29 UTC by Alyssa Manus.
Phases and Fluxes and Functions, Oh My!
In the table below, you can see the corresponding data from the images above. Notice that the measured fluxes Afterglow calculated for the Virgo A images are very similar, demonstrating its utility as a calibration source. To calibrate the flux measurements, the measured flux of the moon was divided by the corresponding measured Virgo A flux and then multiplied by the 229.2, which is the average flux density of Virgo A.
| Name | Date (UTC) | Moon Phase | [Measured] Flux | Calibrated Flux* (Jy) |
| Moon1 | 1/12/2023 21:44:38 | 61% | 3862.698 | 837.6 |
| Virgo A1 | 1/13/2023 09:01:29 | 61% | 1,057.047 | 229.2 |
| Moon2 | 1/17/2023 11:54:29 | 21% | 4600.333 | 997.0 |
| Virgo A2 | 1/17/2023 06:14:14 | 21% | 1,057.745 | 229.2 |
As is evident in the numbers, the flux of the moon increased as its phase decreased, which means the moon got brighter in the radio spectra, and since radio spectra emissions are signs of thermal activity, the data collected can be used to calculate the surface temperature of the moon. The equation to do so appears complicated, but is rather quite simple: to calculate the surface temperature of an assumedly spherical object in Kelvin from a measurement in Janskies, you must divide the flux density converted to W/m2/Hz by 2 times Boltzmann’s constant (1.380649 ⋅ 10-23), multiply that by the square of the speed of light divided by the central frequency in Hz, then multiply all of that by the distance in kilometers to the object squared divided by the pi times the radius squared. For this dataset, the distance to the moon was found using the desktop application Stellarium.
| Name | Date (UTC) | Moon Phase | Calibrated Flux (Jy) | Distance (km) | Temp. (K)* |
| Moon1 | 1/12/2023 21:44:38 | 61% | 837.6 | 403046.501 | 240.5 |
| Moon2 | 1/17/2023 11:54:29 | 21% | 229.2 | 369608.606 | 240.8 |
For greater context, the temperature in Kelvin above has been converted to Celsius and Fahrenheit in the table below.
| Name | Temp. (K) | Temp. (C)* | Temp. (F)† |
| Moon1 | 240.5 | -32.7 | -26.8 |
| Moon2 | 240.8 | -32.4 | -26.2 |
†Calculated by (T – 273.15)(9 / 5) + 32, where T is the temperature in Kelvin.
Baby, It’s cold up there
As the table above indicates, the temperature of the moon does not change with with the lunar phases, which, as hinted to in the sections above, means that the moon has a somewhat constant radio brightness despite having an oscillating visible brightness. The researchers believe the precision of their measurements speaks for itself, and they can assure the readers of the measurements’ accuracy with comparisons to other datasets. The graph below was taken from this website and features measurements of the moon’s temperature sampled in the 3GHz range by Akabane (1955), Medd and Broten (1961), Koschenko et al. (1962), and Alekseev et al. (1966); the researchers’ measurements are indicated by the red circles. Despite being higher-than-average measurements, they still fall within the acceptable range for temperatures of the moon. In all, the researchers are very pleased with their work and hope you enjoyed viewing the images and reading about their results.
Credits
Thank you to Andreas Buzan for working with me on this data and providing the first image of the moon in this dataset. Andreas maintains his own blog at tarheels.live/blogandreas/; check it out to see his interpretation of the same images.
As well, thank you to Alyssa Manus for also working with me on this data and contributing the first image of Virgo A. She also runs her own blog at alyssacmanus.wixsite.com/astrophotography.
And finally, thank you to Ruby McGhee for also also working with me on this data and taking the second image of the moon. Her blog can be found at tarheels.live/rubymcghee/.