This is a webpage copy of an article I wrote in January 2024 for the BAA Venus and Mercury section newsletter, Messenger, in early 2024 (Messenger #14). The newsletter is available to members of the BAA, however, to allow a wider readership I have posted the whole article here. The article has been edited slightly to improve readability.
Martin, 1st Oct 2024
Introduction
During the Northern hemisphere’s excellent eastern apparition of Venus in spring of 2020, drawing inspiration from the groundbreaking work of Anthony Wesley and Phil Miles three years earlier, I embarked on my first attempts to image thermal details of the planet’s night-side (NS).
In 2017 Wesley and Miles had imaged the night-side of Venus at 1000nm and followed a persistent brighter ‘hot-spot’ located at a low-lying area near the dark feature, Aphrodite. There was much interest to follow-up on their observations as there was a chance that this brighter area might be associated with volcanic activity on what was traditionally thought of as a volcanically dead planet.
Successfully imaging of NS thermal variations and hence detection of height variations of the surface of the cloud-shrouded planet, requires working in a narrow band of IR light centred around 1000nm – wavelengths where the Venusian atmosphere is almost transparent and you can therefore see the infra-red radiation from the 470°C surface (figure 1). This imaging requires the use of specialist filters in front of the digital video camera. As the night-side signal is very faint, large apertures win out and I called into action my largest telescope, my 444mm f4.6 Newtonian scope, Fossil Light.
My efforts in spring of 2020 were outlined in the BAA’s Mercury and Venus newsletter, Messenger (#04 June 2020), which this article follows on from. You can download and read that Messenger article using the button below.
You can see the combined result from the best Venus NS sessions from 2020 in figure 2. This shows several well-known darker areas corresponding to higher regions of the surface of the night-side.
Challenges for 2023
Any planetary imager attempting to image the night-side of Venus faces an array of familiar and less familiar practical challenges. The main issue is that the thermal radiation from the night-side is relatively weak and can be swamped by the immediately adjacent and much, much brighter dayside. One also has to contend with the wash of light from the twilight sky, particularly if the Sun is not far enough below the horizon.
The need to keep the background twilight levels low means that best imaging often comes when the Sun is below the horizon and this then unfortunately means Venus’s altitude is never that high. Often this low altitude makes seeing conditions less than ideal. Seeing problems are made worse because the dimness of the night-side and low contrast of the variations necessitates much longer frame exposure times than are normally the case with planetary imaging. This can further exacerbate atmospheric smearing issues that rob detail from processed images.
The best conditions for Venus night-side attempts are when steady and transparent skies prevail, during a period when the ecliptic makes a steep angle with the horizon (putting the planet higher in a dark sky) and whilst Venus’ phase is between 15% to 45%. Too small a phase and Venus is too low in a dark sky, but too large and Venus is smaller in size and the larger dayside seriously starts to swamp the night-side. For Northern hemisphere observers the geometry of the morning apparition in September 2023 was very promising and this article particularly describes my attempts to image the Venus night-side during that period.
Post-2020 Review
In 2020 I had used Fossil Light sitting on my home-made equatorial platform for my night-side attempts. The camera which I found gave the best results was my Gen 1 Sony Pregius IMX174-chipped ZWO mono camera. This was used at prime focus (f4.6). The large 5.86um pixels of the IMX174 sensor gathered more IR photons than other cameras I experimented with and the camera seemed to be free of spurious secondary reflections of the much brighter dayside crescent. Such reflection are a major issue for many potential imaging cameras (more on this below). I combined this camera with an Edmunds 1000nm filter with a 25nm bandwidth, purchased with a generous Ridley grant from the BAA and facilitated by Paul Abel and David Arditti.
After the 2020 sessions, it was clear that the biggest improvement in image quality for future attempts would be achieved by use of a more IR sensitive camera. With the ZWO ASI174MM it was difficult to pull real features out of the noisy, low brightness night-side. This was because there was just not enough night-side data to stack during the short sessions; book-ended between when the sky was too bright and when Venus was too low or blocked by trees north-west of my St. Albans garden. A more sensitive camera would gather more photons and improve signal to noise ratio (SNR) of stacks, allowing better rendition of night-side features. A much more sensitive camera might even allow one to select just the very best data rather than stacking 80-90% of the frames. It also held the potential of operating at image-scales closer to the diffraction limit of the telescope and so pulling out finer detail when conditions were good.
My ASI174MM camera was pretty insensitive at 1000nm with a QE of just 2.8% and the camera had a significant read noise level of 3.5e–. Other CMOS based digital video cameras with Sony sensors were available which were lower read noise and which were much more sensitive at the longer wavelengths. One of the potential candidate cameras was the popular ASI462MC, whose colour filters become almost completely transparent beyond 850nm, making it behave like a mono camera and whose absolute QE was 17% at the required wavelengths. The issue recognised in 2020 with this camera, however, and many otherwise promising candidates, was that the degree of overexposure required to see the night-side led to the appearance of multiple bright secondary reflections of the dayside crescent. Some of these secondary reflections unfortunately overlapped the faint night-side portion, seriously degrading the image quality. An example of such sensor multiple day-side reflections, here with an ASI178MM, is shown in figure 3.
Secondary Reflection Issue
Back in 2020 the exact cause of the secondary reflections that ruled out so many potential candidate cameras with good IR sensitivity, was the subject of some speculation. Now it seems clear that the issue is caused by the regular patterning/metallisation of the CMOS sensor array acting as a diffraction grating and the different diffraction orders reflecting back off the front or rear (or both) of the protective cover-slip bonded to the front of the sensor. The spacing of the pattern of multiple reflections of the Venus dayside are just related to the imaging wavelength and the characteristics of the sensor and its coverslip and are essentially independent of scope focal length and f-ratio. The shape and sharpness of these secondary reflections do, however, seem to relate to the f-ratio – being more well-defined at higher imaging f-ratios.
Removing the cover-slip is one proven solution to eliminate the secondary reflections and has been tried recently by Phil Miles. It is not for the faint-hearted though and needs a very steady hand and great patience. Another option, tried by John Boudreau and others, is to get the cover-slip removed commercially and Salvo Technologies in the US offer such a service for $500.
Before 2023 these alternative approaches had not been tried. I took a more basic route to finding a better camera by testing whatever ones I could get my hands on – attempting to understand each candidate’s infra-red sensitivity and the severity of any secondary reflections.
Methodical testing of cameras involved creating an indoor table-top set-up to mimic imaging the night-side of Venus, working at the same long wavelengths and with the target a similar size on the chip to the image of Venus. To achieve this a 0.4mm pinhole illuminated from behind with a quartz halogen lamp was viewed at 1.5m with a 300mm f4.5 Canon lens attached the chosen camera on test. My Edmunds 1000nm/25nm filter was sited between the lens and the camera.
For each camera the exposure and gain for correct (unsaturated) exposure of the pinhole was recorded. Images were then taken with exposures 100x longer than this, in order to mimic a 100x over-exposure of the Venus dayside. The composite below shows all cameras tested via this method up to April 2023 and shows the severity of the secondary reflections. An image of Venus at the same mm scale on the chip is shown for comparison. Several rounds of testing were completed in 2022 and 2023, assessing several candidate cameras, but none seemed to be as good as the ASI174MM for secondary reflections. The breakthrough was the suggestion by UK planetary imager, Tom Williams, of testing the colour Uranus-C camera (IMX585 sensor) after removing its protection glass. This window was sited in the camera body several millimetres in front of the sensor and should not to be confused with the cover-slip which is bonded to the front of the sensor. With the protection glass removed the obvious halo around the over-exposed pinhole all but disappeared, leaving a large clear space between the dayside and the first set of secondary reflections. These reflections were still present with this camera, but were sited much further out from the centre compared to those on other cameras tested. Note that the ASI482 tested also had secondary reflections well-spaced from the centre of the field and would have been a possible candidate camera, but the QE at 1000nm was only 11% – 2/3 that of the IMX585 chipped Uranus-C.
By measuring the exposure time and gain for correct exposure of the pinhole, and taking account of the different sensor’s relative pixel size and well depth[1], the true relative sensitivity of the tested cameras could be determined at ~1000nm. Knowing the relative sensitivities of the different cameras and also knowing the absolute QE for the IMX462 sensor, one of the cameras tested and which had published data, allowed the absolute QEs of all the cameras to then be calculated and tabulated (see Table 1).
[1] Both these influence the grey level brightness for a given sensor surface brightness and must be taken into account for meaningful QE & true sensitivity calculations. For example, for two cameras with same QE and well depth, but with pixel areas differing in area by 4x, the larger pixelled camera will seem to be 4x more sensitive as the image will be 4x brighter. Conversely, for two cameras with the same QE and pixel area, but where the well depth of one is 3x the depth of the other, it will appear to be 3x less sensitive as the image will be 3x fainter.
As well as showing promise as far as secondary reflections are concerned, the table shows that the colour Uranus-C also had a good absolute QE at 1000nm, being over 5x better than the ASI174MM, after correcting the measured exposure for differences in pixel size and well depth. Like the ASI462MC, the Uranus-C is also a colour camera with Bayer colour filters which become almost completely transparent beyond 850nm, making it behave essentially as a mono camera.
With a much more promising camera for the 2023 night-side imaging sessions settled upon, attention turned to the choice of suitable filters to use with it.
Night-side filters
In 2020 I had used an Edmunds 1000nm filter with a 25nm FWHM (full width half maximum). As you see below, the centre of its bandpass is rather offset from the centre of the VIMS transmission peak taken from figure 1:
Another popular Venus night-side filter choice, championed by Anthony and Phil, in their Venus NS imaging attempts, was a stack of the Thorlabs FELH1000 and the Semrock 935nm/170nm filters. This is an expensive combination, however, at ~£600. The combination is more closely coincident with the peak VIMS transmission than the Edmunds 1000/25 and is chosen to maximise the contrast by actually blocking the 960nm to 1000nm part of the VIMS spectrum.
I discussed suitable Venus night-side filters with Tom Williams in March of 2023 and he found a relatively low cost filter centred on 1010nm with a 38nm bandwidth from PixelTeq at Salvo Technologies. The filter looked really promising for Venus NS imaging, being a much better match with the VIMS spectrum than the 1000/25 Edmunds (figure 7).
The PixelTeq filters are cut from a master disc of 1.0mm glass filter sheet and you can state your required diameter. I ordered one for Tom and one for me, both at 25mm diameter, and mounted them old 1.25″ colour filter housings.
With the test pinhole set-up and using the Uranus-C camera (with its protection glass removed) the 1010nm/38nm filter gave a pinhole brightness of about 150% of that given by the 1000nm/25nm filter. Although I never did any direct head-to-head testing of the Edmunds 1000nm/25nm against the PixelTeq 1010nm/38nm on Venus itself, I do believe the newer filter gave a significant improvement over the Edmunds in terms of image brightness and contrast.
A filter improvement that I introduced between the evening and morning apparitions in 2023, was too add an 850nm ZWO filter in series with the PixelTeq filter. This addition reduced light leakage in the visual portion of the spectrum. The leakage could be readily picked up by viewing an incandescent bulb by eye through the PixelTeq alone and this disappeared when viewing with the ZWO 850nm filter in series with it. This leakage manifested itself as a background Bayer pattern when imaging the Venus NS in bright twilight without the ZWO filter.
Another cause of light leakage which could result in a background Bayer pattern was discovered by Anthony and Phil during the evening apparition. The cause of this was significant light leakage through the data sockets in the body of the Uranus-C. The simple remedy was to replaced the original thick black foam sealing ring that encircles the sensor, and which had previously been removed when the protection glass was taken out.
Image Scale
The Uranus-C had 2.9um pixels compared to 5.86um for the ASI174MM. This meant that pixels of the Uranus-C would have less than 25% of the area of the ASI174MM pixels. Despite this, the 5x improved QE of the Uranus-C combined with the better fit of the PixelTeq transmission spectrum with the VIMS spectrum and its wider bandwidth should have given a significant improvement in SNR for the proposed new set-up. The difference in pixel size would also improve the image scale at prime focus, however, moving things closer to satisfying the Nyquist criterion, the image scale at which resolution would potentially be maximised.
To satisfy the Nyquist Criterion you need an image scale where 2 to 3 pixels cover the smallest (diffraction related) feature a scope can produce – normally said to be features at Dawes limit. A simple rule of thumb encapsulating all the relevant Nyquist parameters, has this achieved for a mono camera when the f-ratio is about 3x to 5x the pixel size in microns. For 2.9um pixels that calculation gives best operation at f9 to f14.5. However, this rule of thumb applies for green light at 550nm; at 1000nm the size of the diffraction limit is almost twice as large. This means that for the Uranus-C, Nyquist would be satisfied at just f5.5 to f8. When operating at f4.5 I am under-sampling just a little. At the next apparition, the good evening apparition for the those in the northerly latitudes, in Spring 2025, I am hoping to try a 1.5x Barlow when the planet size is small enough that the night-side won’t overlap with the first set of secondary reflections when the Barlow is in place.
Groups IO Forum
In late April 2023, in response to an increasing interest in imaging both the day-side and the night-side of Venus, Anthony Wesley set up a Groups IO discussion forum to encourage the exchange of ideas and information. Early members and key early contributors, in addition to Anthony, included Peter Tickner, Chris Hooker, Tom Williams, Phil Miles, Niall MacNeill, Luigi Morrone, Javier Peralta, Manos Kardasis, António José Saraiva Da Cunha Cidadão, and myself. Since then others have joined and by the end of December 2023 the forum had 23 members.
The forum has been a tremendous success with well over 300 topics discussed in the first 9 months, peaking around the periods before and after inferior conjunction (IC), when conditions were best for imaging the Venus NS .
Personally I have found the forum invaluable for the exchange of information and as a sounding board for new ideas, as well as motivating me to get up early and image during the morning apparition! It has a strongly positive and encouraging culture, and has really helped to move things on as far as amateur imaging of the Venusian night-side is concerned.
Example topics discussed during 2023 have included;
- Latest night-side image results from Venus NS imagers across the world
- Latest results from different cameras – particularly secondary reflection issues
- Causes of secondary reflections in night-side imaging particularly in sensor cover-slip and arising from reflections of filters
- Latest on different NS filter and filter combinations
- Visual leakage through 1000nm filters
- Experimentation on apodising masks to suppress secondary vane issues
I would strongly recommend anyone interested in imaging Venus, particularly the night-side, to consider joining the forum at: https://groups.io/g/VenusImaging.
Volcanism on Venus – NASA Findings
In mid-March 2023, with the night-side season fast approaching, NASA broke some big news. Recent analysis of 30 year-old Magellan probe images by a team led by Robert Herrick, found incontrovertible evidence of recent lava flows on Maart Mons, in the Alta Regio region of the planet.
The news that Venus is indeed still volcanically active does not prove that Wesley and Miles’ bright spot near Aphrodite was definitely volcanic in origin, but does strengthen the case somewhat, especially as the bright spots was there again in several images in early 2022. The news gave significant added impetus to those gearing up to imaging the Venusian night-side in summer 2023.
June/July 2023 Imaging Results
Settling on the Uranus-C camera with the PixelTeq 1010nm/38nm filter on my 444mm Newtonian at f4.5, would, I thought, give me the best chance of picking out detail on the night-side and improving on my best attempts from 2020.
I started my night-side campaign on 4th June 2023. That evening, with the phase at almost exactly 50%, I was just able to tease out the night-side from glare from the overwhelmingly bright dayside.
In total I imaged the night-side on seven occasions before inferior conjunction, the best being on the 23rd June with the phase at 38% and an altitude of 14°, when some of the more prominent features were visible (see figure 8). The 19th was the only other pre-IC session where darker features were imaged. That night I picked up, Alta Regio, near the limb.
The pre-IC sessions were never going to be very good, given the poor geometry for northern latitude imagers which meant that Venus would be pretty low in the sky around sunset.
The pre-IC sessions, however, did allow me to test my new set-up and to try out different imaging settings before the promising post-IC period. The camera and filter combination I had chosen for NS imaging worked well with significantly increased brightness and improved SNR compared to 2020.
The last session of the pre IC set was on 7th July when the phase was 27% but the altitude when imaging was only 10°.
Apodising Mask experiments.
During the evening apparition my secondary vane orientation was such that one of the wide vane diffraction spikes from the secondary vane went right through the NS region on all my images. This adversely affected the contrast of any real surface features as you can see in the example shown in figure 8. In 2020 I had been luckier with the geometry – the night-side sitting nicely between two diffraction spikes.
To effect a solution, Chris Hooker suggested I tried an apodising mask over two of the vanes to redirect the diffracted light away from the night-side. Following his advice I tried a number of different apodising mask configurations. The masks were made of opaque aluminium foil on PET, cut on an XY cutting plotter at work. The most successful design can be seen attached to the front edge of the two offending vanes in figure 9.
With some designs of apodising mask the vane diffraction situation was made much worse, as you see in figure 10 where overlapping secondary images were created.
With the right apodising mask in place, however, the wide diffraction spike going through the dark side was indeed reduced as the images below in figure 11 show. The mask used was a 30° saw-tooth on a 3mm pitch, shown in place on the vanes in figure 9.
Eventually I ran out of sessions before the mask design could be fully optimised and I just hoped that in the morning sessions, where the apparition geometry would be different, I would be as lucky with the placement of the spikes as I was in 2020. As it turned out, luck was on my side in the September sessions and an apodising mask was not needed.
The topic of apodising masks for Newtonian telescopes for NS imaging is work in progress, possibly to be picked up again at some future date.
September 2023 Imaging
The June/July pre-IC sessions were the warm-up for the much more favourable and greatly anticipated September post-IC morning appearance. Then the geometry of the early part of the apparition was going to be much better. The morning sessions, however, would mean capturing Venus at a place in the sky where I’d never previously imaged and I needed to plan carefully for it. The only viable location in my garden for Fossil Light to access this part of the sky whilst on its equatorial tracking platform was at a location on the lawn under a lowish greengage tree – a most unconventional location! The scope would have to look over my garage roof, looking into a gap between my roof and my neighbour’s roof.
I did an evening test run with the proposed set-up and it all seemed quite feasible to image the part of the dawn sky where Venus was predicted to be. The hardest part was assembling my 2.3m tall scope under a tree canopy that started at a height of just 2.0m above the ground – a special tail prop was designed and all systems were go.
After two years of anticipation and several sessions of planning and experimentation it was crunch time! My first morning session was on Tuesday 5th September when the phase was at 15% and the planet 47″ diameter.
Getting up at 5am does not come easily to me, but the excitement got me over the tiredness. I had set-up the scope at the unusual location under the greengage tree the evening before and covered it before going to bed. On rising all I had to do was pull the tarpaulin off and wait for Venus to rise clear of the garage roof. The sky was cloudless and the transparency good. Eventually the view of Venus was unobstructed and I managed to capture 16 precious night-side videos each 2 mins long.
During the session I tried a range of exposures from 75msec to 150msec, which was ~40x to 80x overexposed for the dayside. Gain was set moderately high at 300 (30dB) and 240 (24dB) to keep the read noise low. As usual I captured at maximum bit depth; this was the ’16-bit’ setting in Firecapture which actually gives 12-bit depth.
That morning I continued to image until the Sun rose and the dawn sky was too bright. Everything had gone well with no hitches, the seeing was good and I had lots of wonderful data to process that evening.
The processed results from this first session were very pleasing with plenty of features visible in the night-side – so much better than any of the pre-IC sessions and indeed better than any of my single sessions back in 2020.
The following morning the forecast was also good, so I imaged again. The sky was cloud-free but badly affected by Saharan dust causing dimming of about 3.5x and forcing me to push exposures up to 300msec. To my surprise, results turned out even better than the day before. The final stack was a bit noisier but the seeing had actually been significantly better than the day before, rendering finer details in the NS.
Sky Brightness
During the optimum imaging period in 2020 I had to image the Venus night-side in quite a bright sky. This was due to trees to the NW blocking me from imaging Venus when the sky was darker, later on in each of the sessions. The high level of shot noise from the wash of bright twilight which had to be subtracted off the night-side, left its imprint on the NS region and the resulting signal to noise ratio (SNR) was quite poor. These were the sort of issues that deep sky imagers face when capturing DSOs under badly light polluted skies, not the sort of problems normally faced by planetary imager. It was a big learning experience for me trying to fathom what was going on.
Back in 2020 I took measurements of sky brightness and determined that once the Sun was more than ~5° below the horizon, sky brightness issues with NS imaging waned significantly. The sky could be still bright to the naked eye but that didn’t matter much. As the Sun rose from the minus 5° elevation, however, the sky brightened dramatically – being some 50x brighter when the Sun was on the horizon compared to that at minus 5°.
These morning sessions in September were wholly different to 2020. I was able to image with Venus at a reasonable altitude, unobstructed, in a sufficiently dark sky leading to dramatically improved SNR in the images.
Early on in Sept I assembled an image montage based on data from the 5th, comparing the appearance of the Venus NS as a function of solar altitude, from -5° to -1.1°. This is shown below in figures 15a and 15b. All images were 2 min videos with the same accumulated exposure time and processed with the same wavelets scheme. There was some variation in gain and exposure times, as indicated, but not enough to undermine the thrust of what the montage shows, which is the increasing noise imprint of the sky as the solar altitude increases. As the Sun approaches the horizon, dust motes make their appearance too and in the last image an 8-pixel grid characteristic of the IMX585 sensor in the Uranus-C is seen in its full glory.
Processing
The Autostakkert3! quality sort, align and stack settings used on the data from these imaging sessions were fairly simple, but did involve stacking a high proportion of frames (90% generally) to keep the SNR high, and also using just a single alignment box encompassing the day-side and night-side.
Registax wavelet settings when processing NS data needed to keep glare from the day-side under control whilst improving the contrast of the relatively large surface features. With lots of good data to play with from the sessions of the 5th and 6th Sept, and other imagers to discuss things with on the Groups IO forum, I was able to experiment and find improved Registax wavelet settings compared to the ones I had used in 2020 and indeed for the earlier pre-IC data. These new settings, which evolved from playing around with a set published on the forum by prolific UK imager Peter Tickner, worked well, though they were well outside the normal values I would have use for regular planetary processing.
For the record, the good settings I found for the wavelet sliders are shown in figures 16a to 16d below. I dispensed with the increased gamma I used in 2020 to control the day-side, keeping it at unity. Like then, however, I did use the histogram stretch in Registax to boost the night-side and subtract off any wash of twilight across the frame. This was done by setting the low histogram level at the start of the histogram and setting the high level at the start of the main hump as shown in the progression of figures below.
Later Sessions
By Sept. 9th the phase had grown to 18.5%. There was high cloud around but the seeing was fair and some good details were picked up. A faint brighter spot was seen at the same location where Wesley and Miles saw their hot-spot in 2017 and this is arrowed in figure 17.
On the morning of the 15th Sept. Venus was considerably higher in the sky when I started that imaging session compared to the planet’s elevation at the start of the sessions at the beginning of September. Conditions were near perfect, with very good transparency and good seeing, significantly benefitting from the higher altitude which arose from the increased solar elongation.
I was able to take about twenty 2min. videos all with the Sun more than 5° below the horizon. I had by this session settled on an exposure of 75msec and gain of 260 with the Uranus-C camera as it had worked well. The session yielded the best images of the apparition and really underlined the importance of good seeing when other issues have been largely overcome. Oddly my first processing showed some weird regular banding across the NS looking like Newton’s rings, but this was removed using the FFT edit feature in Image-J, which can very effectively remove regular patterning when done carefully.
My last decent session before the weather inevitably deteriorated was on the 23rd of September. Seeing was on a par with that on 5th of September and I had a good view of the whole of Aphrodite with the phase now at 34.5% (figure 19).
Gain and Exposure Comparison
During this longer session I finally had time to run a set of comparison videos that I’d wanted to do, with a matrix of different exposure times and gains.
I did one set of three images at fixed gain and three different exposure times: 25msec, 75msec, 225msec. The other set I used the same three exposure times but altered the gain up or down by a factor of three to maintain the same image brightness – gain went down by 3x between each video as exposure went up by 3x.
In processing each 2 min video of the comparison I chose the best 90% of each and processed them identically in AS3! and Registax, barring the histogram stretch in Registax which was tailored to match the different brightness of each stacked image. No other processing was used.
The processed results are shown in figure 20.
Examination of the comparison composite shows much less of a difference as a result of changed capture settings, than I had expected. The following was seen:
- in these conditions the 25msec images showed the most detail, probably because the shorter exposure reduced atmospheric smearing
- the 25msec and gain 260 image suffered the most from read noise as the signal is lower in comparison with the fixed read noise in each frame. Although the short exposure means there are more images to average it out it never fully makes up for the lower signal.
- the 25msec gain 370 image shows less background noise than the 25msec 260 gain image as the read noise is slightly lower at 370 than 260. This marginally the best image.
- for the wavelet settings used, the day-side of the 225msec image at gain 260 (top row RH end) hides a similar amount of the night-side as the dark (wavelet induced) band around the day-side does on the 25msec image at gain 260 (top row LH end). You might imagine that by dropping the exposure you might see more of the night-side as the degree of over-exposure of the dayside would be less, but this was not the case.
- in poorish seeing there is an advantage in reducing the exposure time but keeping the gain highish
Map of Venus NS
As in 2020 I was again able to use the map computation function in Winjupos to create a rectilinear map of the Venusian NS using images spanning my Sept. imaging period. Each day the planet’s central meridian advances by 2° and this means that during the optimum 3 weeks of so for imaging the night-side, the view advances by some 40°. In the map below I have combined my 2020 evening apparition results with my 2023 morning results to cover a wider area of the planet. You can clearly seen the superiority of my 2023 results to the set from 2020 with much more detail picked up.
As the planetary mechanics of our planet and Venus mean that essentially each morning apparition of Venus we see the same night-side face and each evening apparition we see a fixed different face, I am fated to never get to fill in the area between longitude 120° and 190° on my map! Having said that, I do hope to be able to update the evening apparition strips with better ones during the good apparition in Spring of 2025.
The most detailed area of the map in figure 21 is the strip associated with the 15th Sept. image. On the map below (figure 22a) I have highlighted the most detailed region of that strip. This is an area from 30° to 75° longitude and +45° to -15° latitude:
For this rectangular area I have below created a side-by-side comparison between my night-side data and the Magellan altimetry data, labelling some of the elevated areas with their corresponding names. There seems to be a reasonable match between the two sets of data, especially in the upper half of the rectangle.
Streaks and Bright Spots
One of the key motivations for imaging the Venus night-side was to see if there were any persistent bright spots that were visible for the whole of the apparition and which rotated with the planet. Such spots might be evidence of ongoing volcanism on the planet’s surface.
In my images the only brighter spot seen was in the image of 9th Sept (figure 17). The location of this spot coincided with the location of the persistent light spot seen by Wesley and Miles near Aphrodite in 2017 and imaged again by them on several occasions in early 2022.
This spot, although seen in my image from the 9th, was not seen in the much higher resolution image taken in the good seeing just 6 days later, on the 15th Sept. A brighter spot at this same location can also be well seen in Anthony Wesley and Phil Miles’ image of 12th (figure 23a) and especially the 15th (figure 23b) but not in their images of 11th, 13th 14th or 16th. This transient nature suggests this bright spot is not a hot volcanic surface feature – like a lava flow or eruption. What is causing this area to appear brighter/hotter is not known.
During the apparition, several imagers occasionally reported imaging brighter streaks across the night-side. Such a streak can be seen at the very bottom of the night-side in my image from the 15th (figure 23c).
Wesley and Miles image from 15th Sept., taken 14.5 hrs later (figure 23b) also shows a bright streak, similar in size and shape to my image, but further north. By 16th Sept. there is a curled brighter streak closer to the limb, similar in length to the feature in the other images from 15th.
These streaks also seem to be transient features whose lifetime is short or which move rapidly across the disc. Whether the streak seen in the two images from 15th and the one from the 16th is the same streak drifting in position is not known. The study of such streaks does seem to be one worthy of attention at future apparitions to try and determine their true nature. Anyone interested in carrying out such a study should be aware of the tremendous portfolio of images that Anthony and Phil have amassed from this and previous apparitions – a collection that may already hold much useful evidence.
Concluding remarks from 2023
This article has been considerably longer than intended but I do hope you have found some interest in hearing about my Venus night-side imaging journey.
You can read more about imaging the night-side of Venus, particularly information about 16-bit imaging, stacking and read noise, which I haven’t had the space to go into here, by downloading a copy of the Venus night-side presentation that I gave at the BAA Winchester weekend in April 2023. A link to this presentation can be found at the bottom of my webpage https://skyinspector.co.uk/my-articles-talks/
To whet your appetite for the next decent imaging opportunity, I’ll leave you with Martin Panell’s simulation for 55°N for the evening apparition in 2024/2025 from www.nakedeyeplanets.com
The geometry for night-side imaging should again be excellent for Northern hemisphere imagers in Feb/March of 2025.
Here’s hoping that this article, together with the anticipation of the good apparition in 2025, will enthuse others to have a go at this most enticing of planetary imaging challenges.
Follow-on Article on Venus NS SNR. Messenger #15 – July 2024
A follow-on article I wrote about measurements of the signal to noise value (SNR) for Venus NS images, was published in the BAA Mercury and Venus newsletter, Messenger (#15), in summer of 2024. You can download a pdf copy of that article by clicking on the button below: