# How close are we to observing all of the sky all of the time?

I was musing on how amateur astronomy is still a heavy contributor to the overall field, and how on any given night there will be plenty of small telescopes pointing into the void and capturing the sort of chance stuff the large telescopes never see because they are observing something specific elsewhere. And this made me wonder, with the combined set of professional and amateur telescopes, do we come close to seeing all of the sky all of the time?

Another way of phrasing the spirit of the question is: If some transient event occurs over the course of seconds, and is not visible to the naked eye, approximately how likely would we (meaning humans) be to notice it?

• I'd be surprised if the answer isn't "we're already doing it". The real question is how much detail we have of the sky all of the time, i.e. how luminous can a transient be and still be undetected. Commented Jan 27, 2022 at 9:56
• One might also ask what is "all the time"? How long can the time be between subsequent images to still count as "all the time"? A normal camera has a dead time between two subsequent images and you need two synced cameras with an offset of half their acquisition time to really have a continuous data feed. Commented Jan 27, 2022 at 10:10
• Surveyed depth and number of filters/passbands measured matter for the answer. NEO surveys e.g. ATLAS or facilities like ZTF already cover the visible sky each night but only to a relatively shallow depth or in a single filter (which normally prevents classification of static transients) Commented Jan 28, 2022 at 1:09
• all of the sky all of the time isn't possible, as far as I understand: we can never see what's on the other side of the sun from us until we move around in our orbit Commented Jan 28, 2022 at 9:49
• On the ameteur astronomy angle of the question - which the answers don't really tackle - the largest near earth asteroid discovered so far this year, 2022 CO6, has just been discovered by amateurs. That's right - the largest one, magnitude 25.6. That's larger than the Chelyabinsk asteroid which caused damage to1000s of buildings and associated injuries. In addition - they discovered it a 5 days before close approach - the most warning time any survey has given for a close approach this year so far. Well done MAP (Alain M aury, Georges A ttard and Daniel P arrott). Très bien! Commented Feb 15, 2022 at 8:42

To answer the first phrasing of the question: "Not yet". Approximate answer to the second phrasing below, after discussion. Adding to GrapefruitIsAwesome's answer, there are a couple of problems in addition to Daylight and Cloud Cover:

• Moonlight: Once every 28 days, the moon is full and makes the whole sky glow (in the same way that the sun does during the day, but to a much lesser extent). As this makes it impossible to see faint objects, all surveys shut down for 3-7 days for this period (depending on the survey and what they consider to be faint), leaving the sky more or less unmonitored. Let's say this means we only see around 85% of what we would otherwise see.

• Opposition Effect: For non luminous objects, the amount of light received from them by an Earth based optical telescope is heavily influenced by their solar elongation (angle between the Sun, the object and the observer). I.e. even if it is night and we are looking at the right part of the sky as an asteroid whizzes by, we get to see very little of the sunlight reflecting off it. This means there is in fact a very limited part of the sky where we manage to observe non-luminous objects. Because of this, over half of the asteroid discoveries we make are in a narrow cone facing directly away from the sun, which covers less than 4% of the sky. See Wikipedia. Let's say this means we only see around 8% of non luminous objects we would otherwise see.

In fact these issues are related. The moon is so bright when it is full because it is in opposition, and significantly dimmer the rest of the time. However, the Sun is much brighter than the moon, and never dims, so Daylight is definitely a bigger problem. It's easy to forget how big an issue this is.

Because we see the day/night cycle from the spinning earth it's easy to think "I only need to wait a few hours to see that bit of the sky". Unfortunately of course it's you that's moved. The daylight stays in roughly the same direction in the sky. To see that part of the universe you need to wait months, not hours for it to be night... and the bit of space between Earth and the sun, you never get to see with an ordinary telescope*. Let's say this means we only see around 50% of what we would otherwise see.

So in an attempt to give an approximate answer to "how likely would we be to notice it":

• For luminous objects: P(not full moon) 0.85 x P(not in daylight) 0.5 = ~43%
• For non luminous objects: P(not full moon) 0.85 x P(illuminated enough) 0.08 = ~4%

Note these approximate numbers ignore cloud cover, airglow and various other lesser factors, and additionally limited by the sensitivity of current telescopes.

* For this reason there's a space telescope being built (NEOSM) that will orbit at Sun-Earth L1, the point at which Earth and the Sun's gravity cancel out. It will look back towards Earth with the Sun behind it, so that we can at least see into the part of space between it's orbit and the Earth. For reference it's about 1% closer to the sun than we are, which may not sound like much, but that's still about 4 times further than the moon. That's the LaGrange point on the opposite side of Earth to the one that Webb is currently orbiting, L2. Both are around 1.5 million km from Earth

• worth noting that you can't solve this from Earth of course. Space telescopes have a much better view. Being in space means blue sky isn't in the way, but they still can't look too close to the sun or the instruments get fried. To see the whole sky the whole time, you really need telescopes orbiting the Sun... however the data backhaul then becomes a problem - the further the distance from Earth, the lower the bandwidth available to send back what they've seen. Tricky! Commented Jan 29, 2022 at 13:13

If we limit ourselves to defining "all of the time" to at least once per day while it is night to avoid the issue raised by Aaron F of only observing optically when the sun isn't in the sky (avoids day time and polar regions with 24 hours sunlight) there are couple that I was able to quickly identify:

1. Zwicky Transient Facility (ZTF) (Thank you astrosnapper)

The Zwicky Transient Facility (ZTF) is a public-private partnership aimed at a systematic study of the optical night sky. Using an extremely wide-field of view camera, ZTF scans the entire Northern sky every two days.

1. Asteroid Terrestrial-impact Last Alert System (ATLAS) (again astrosnapper)

ATLAS is an asteroid impact early warning system developed by the University of Hawaii and funded by NASA. It consists of four telescopes (Hawaii ×2, Chile, South Africa), which automatically scan the whole sky several times every night looking for moving objects.

1. The All-Sky Automated Survey for Supernovae (ASAS-SN)

We are changing that with our "All-Sky Automated Survey for Supernovae" (ASAS-SN) project, which is now automatically surveying the entire visible sky every night down to about 18th magnitude, more than 50,000 times deeper than human eye.

There are probably more, so I think we do cover all of the visible sky possible several times per day, depending on our definitions.

A couple of thoughts:

Daylight

From a given site you are limited to surveying optical only in darkness so you won't have coverage during the daytime. This is only worsened in polar regions during local summer (with a corresponding benefit in winter).

Cloud Cover

You can't survey when it is cloudy, but if you have multiple geographic sites one location may be clear while another is clouded out. This redundancy also helps with issue like failure and maintenance time.

• See also Kiso observatory. Scans the visible sky in 2 hours: en.wikipedia.org/wiki/… Commented Jan 29, 2022 at 11:50