In Part 1, we looked at light pollution and its effects on broadband and narrowband images, and the filters that are most effective at reducing those effects.
In Part 2, we’ll look at imaging techniques, camera settings, and software processes that help further mitigate the effects of light pollution on deep-sky images.
You can’t always move to a less light polluted site, but there are things you can do that to reduce the light pollution you.
Target Selection: A brighter target that stands out against the background glow of the night sky will yield the best results. And targets that can use narrowband filters will always fare better than broadband targets.
Target Altitude: The higher the object in the sky, the farther it will be from light sources on the ground. Choose targets that are higher during your imaging sessions. Stop imaging when the target gets too low and move to another target that is higher up. That might be at 45 or 50 degrees above the horizon, but it depends on the light pollution you’re dealing with. The worse it is, the higher you’ll have to do your imaging.
Moonlight: Try to avoid imaging when the moon is bright, as that just adds extra unwanted light. Broadband imaging should be avoided in moonlight, but Hydrogen Alpha data can still be captured if the target isn’t too close to the moon. Oxygen III data is highly subject to the effects of moonlight, though.
Stray light from nearby lights (also known as light trespass) is another type of light pollution. For example, you might have a dark suburban backyard where a neighbor’s bright patio intrudes.
The best solution is to talk to your neighbor and suggest that they use a shielded light fixture that keeps the light in their yard. See the International Dark Sky Association’s website for helpful advice. Otherwise, blocking the light source with an observatory wall or dome, building a fence, or planting bushes might be helpful. There are also several things you can do with your observatory or telescope to reduce the stray light that gets into your camera.
Dew Shield: This is a useful device that effectively blocks unwanted light from angles away from the direction of the target. Almost all telescope designs will benefit from adding a dew shield. Truss telescopes can benefit from fabric shrouds covering the open parts of the tube.
Flocking / Painting: Stary light can reflect inside the telescope and create artifacts in the image. Shiny telescope tubes can be flocked or painted with very dark black paint. Flocking can prove difficult to use at times, and the “fuzzy” dust needs to be kept off the optics.
Of the black paints I tried, by far the darkest was Musou Black. It’s almost magically effective at darkening telescope tubes, focusers, and extension tubes.
CMOS Camera Settings
With the shift towards CMOS cameras, imagers can now adjust the gain of their cameras, which alters the relationship between photons and pixel values. This relationship is complex, as it changes the full well capacity of the pixels (how much light they can hold before saturating), the read noise, and the dynamic range (the ability to capture the difference between very light and very dark).
When imaging from light polluted skies, the read noise of the chip is less critical. The small source of unwanted noise in the image caused by the read-out electronics is insignificant compared to the “noise” of the light polluted sky background we’re dealing with. The read noise can be reduced by increasing the gain, but that will also lower the full well capacity.
My recommendation is to use lower gain settings to increase the full well capacity and dynamic range, which are most important when capturing data from light polluted skies. The slight increase in read noise is well worth the extra headroom to ensure the brighter areas of the image are not washed out.
Sub-exposures (commonly called “subs”) are the individual exposures that are added together to create your final image through a process known as “stacking.” When imaging from heavily light pollution skies, the sub-exposures will need to be short to avoid saturating the image with high level of incoming light coming (target and background).
Total Integration Time
The total integration time is the sum of the individual sub-exposures you’ve captured. In this case, you need to increase the total exposure time. With light polluted skies, the signal to noise ratio is lower – the target doesn’t stand out from the bright background as well – so you need more total integration time compared to imaging from a dark site.
How much extra time is needed depends on the light pollution level – the brighter the sky background, the more total integration time is needed. At some point, the background light pollution becomes too much and no amount of integration will suffice.
The image below compares a short integration time of three hours to the same target with a total integration time 21 hours. The extra signal to noise from the longer integration time ratio allows more processing that can extract more details, colors, and contrast.
It’s important to calibrate astronomical image. Eliminating uneven illumination with good flat calibration is essential when dealing with the impact of significant light pollution. Once flat-calibrated, you can confidently remove the background glow and gradient caused by light pollution knowing you’re not removing faint detail from the target.
The example below is a luminance frame taken form a Bortle 5-6 location with and without flat field calibration.
Pixinsight Processing – NSG
Normalize Scale Gradient (NSG) is a third-party script available for Pixinsight. NSG examines all the frames and adjusts the brightness and gradients of the frames to match a reference frame. The frame with the lowest background gradient (which often has the highest signal-to-noise ratio) is designated as the reference frame.
Once the script adjusts all the frames to match the reference frame, the data is integrated (stacked) into a final single image, often resulting in better contrast.
This process is CPU intensive, and the settings may need to be fine-tuned, but I have found NSG to be an effective tool for reducing the background gradient caused by light pollution.
The example below illustrates the effects on a dataset integrated with and without the use of NSG.
Pixinsight Processing – DBE
Another Pixinsight tool that is very effective in removing background glow and gradients from a single image is Dynamic Background Extraction (DBE). DBE can be used in conjunction with NSG by applying it to the single image that was enhanced by NSG.
The user has to generate background sample points and adjust the settings on each image for DBE to work best (a process beyond the scope of this guide).
DBE is very powerful, sometimes even transforming a light polluted image into one that is devoid of background gradients and presenting a great deal of faint detail. I highly recommend mastering this process.
Imaging from light polluted sites can be a challenge but imagers have a number of tools at their disposal to optimize their results.
Choosing the right camera and filters is important but you still need to know the most effective camera and exposure settings, and other imaging techniques.
Most important of all are the processing techniques that allow you to extract the information present in your images that’s masked by light pollution and render aesthetically pleasing results.
It’s unfortunate that light pollution has robbed many of us of the natural night sky. But we’re also very fortunate to have the many improvements in equipment and software that allow us to deal with that ever-growing problem.
Rouzbeh Bidshahri has many examples of images taken from light polluted skies with a monochrome camera in his gallery at www.RouzAstro.com