Finished! Looks like this project is out of data at the moment!

See Results

Thanks to your amazing work we have finished classifying SW-HSC!! You will still be able to continue to classify but you may see more training images than usual.

Education

Lensed Galaxies
Lensed Quasars
Group-scale lenses
Lens Impostors:Disk galaxies
Lens Impostors:Neighbors
Lens Impostors:Loners or Edge-on Disks
Lens Impostors:Lightweights
Ambiguous cases
Stars
Artifacts
Build your own lens

Gravitational Lenses

See the definition of some lensing related terms in the FAQ.

Lensed Galaxies

A strong lens consists of at least two galaxies: one in the foreground, doing the lensing, and one in the background, the image of which is distorted. Lens galaxies are typically very massive, yellow/red in color, with an ellipsoidal shape. Most of the images you'll see feature one such galaxy in the center. A lensed galaxy is usually, but not necessarily, blue, and is often distorted into multiple images that trace a circle around the lens. For each system here, we show two panels - the original image (on the left) and light of the lens-galaxy-subtracted residual image (on the right).


Above is a clear example of a strong lens (first system), recognizable from the big blue arc and the small blue "counter-image", located on the other side of the lens galaxy, opposite to the arc.
The presence of counter-images can help a lot in determining whether a galaxy is a lens or not: look out for them! Most of the lenses we're looking for, however, are not as obvious to spot as the one above. The second system, for instance, is a lens for which only one arc is visible, with no counter-image (the counter-image is there, it's just too faint to show up in our data).


In some cases, such as the example above, the counter-image lies on top of the light coming from the foreground galaxy. The lens light-subtracted image (right panel) can be very helpful in these situations.


When lens and background source are very well aligned, an almost perfect circle is formed (see above).
In these cases, viewing the lens light-subtracted images can be quite handy.

Go to the Top

Lensed Quasars

|
Lensing does not always produce arcs. When the background source is small in size, such as a quasar or a compact galaxy, it can be lensed into sets of point-like images, like the examples above. Four-image systems (called quads, on the left) are quite common. Two-image systems (called doubles, on the right) are even more common, but are not as obvious to identify as a "quad".
Go to the Top

Group scale Lenses

Galaxy groups and clusters are gravitationally bound entities. They are the most massive systems in the Universe. As such these are the most powerful cosmic telescopes and can produce large, spectacular arcs and multiple images.

The above two examples show "group scale lenses" where the foreground lens comprise of a group of galaxies (usually of similar colors). The separation between the multiple arcs produced by group scale lenses is much larger than those seen around individual galaxies because it is the combined mass of the group that contributes in producing the lensing effect. The number of arcs or the arrangement of arcs around the galaxy groups can also be non-standard or complex.
Go to the Top


Lens Impostors

Lens impostors are astronomical objects that look like gravitational lens systems but are physically different. Below are some typical examples of lens impostors that you can familiarize yourself with to aid your classification.

Disk Galaxies

Many galaxies have yellow "bulges" in the centers, surrounded by a disk of gas and blue "spiral arms.". We call these "disk galaxies". Unlike gravitationally lensed arcs, the arms don't trace a rough circle around the bulge and they are usually accompanied by some fuzzy distribution of light. In these examples, you can see fuzzy arms that either don't trace out a well-defined shape, or a hint of a "bar" that connects the arms to the bulge.


Some galaxies are surrounded by a blue belt of young stars. When looking at one of these "ring galaxies" face-on, it can be difficult to distinguish it from an Einstein ring of a gravitational lens. However, because of their random orientation and inclination with respect to the plane of the sky, most ring galaxies appear to be flattened, with the blue region describing an ellipse-like shape. In contrast, the rings produced by gravitational lenses are usually rather round.
Go to the Top

Neighbors

Images taken of the sky are a flattened 2D projection of a 3D universe. This means that physically unrelated and far-away galaxies appear next to each other in the 2D images - this is a chance or random alignment. Sometimes the shapes or patterns of these neighbouring galaxies are such that they can mimic lenses, but there could be clear signs that these are not gravitational lenses. For lenses, remember to look for an arc that is well-defined and curved around a plausible lensing galaxy, and also look for hints of a counter-image. If things look very extended and fuzzy, or curved the wrong way, or too straight, they are most likely not lenses. Some examples follow.

Left: The blue feature above has a very flat and straight shape, which is unusual when the "lens" is a single galaxy as in this example. Real gravitational lenses have arcs which seem to curve around the lens galaxy at the center. It is much more likely that this is a galaxy, somewhere in the line-of-sight that just happens to be elongated next to a yellow galaxy in the foreground.

Right: In the case above, the supposed "arc" is broken into two features. Not only the blue light is too fuzzy, the two blobs do not fall on an arc / or a ring-like pattern combined with the fact that there are no additional images making this a lens impostor.
Go to the Top

Loners or Edge-on Disks

In the images above, there are elongated features that could resemble gravitationally lensed galaxies. However, there are no plausible nearby lensing galaxies, suggesting that they are not lensed arcs. Even if there was a yellowish lensing galaxy in the vicinity, notice the light distribution in these features is concentrated towards the center and tapers towards the edges. This is a sign of a disk galaxy seen edge-on rather than a gravitationally lensed galaxy.

Lightweights

This configuration of multiple blue galaxies on the left looks like this system could be a lens, but the yellow galaxy appears fairly small, implying the mass is also small, which is unlikely to create lensed images that are located so far apart. The two elongated galaxies on the right look like they could be arcs, but there is only a very small galaxy between them, suggesting that these are not lensed images.

See some examples of real lenses to build an intuition or a mental model in order to recognize them correctly.

Go to the Top


Ambiguous cases

In a lot of cases, it's hard to tell whether a galaxy is a lens or not, as in the two examples shown above.
The object on the left shows a blue image very close to a galaxy. The blue feature is slightly elongated and aligned in the same way a lensed arc would be, but it looks boxy rather than arc-like. The boxy shape, however, could be the result of the low resolution of the image.
The image on the right shows some characteristic features of a gravitational lens, such as a nice round arc oriented towards the central galaxy, as well as a fuzzy distribution of light more typical of spiral galaxies.
In cases like these, we would like you to be inclusive and label them as potential lenses.

Go to the Top


Stars

In addition to galaxies, there are also stars from our own galaxy that are visible in our images.

Young stars appear bluish white and older stars appear reddish yellow. See two examples below which show how a reddish star (left) could look different from a reddish galaxy (right). Both have a bluish object near them. The left object should definitely be rejected whereas for object on the right, its worth looking at lens light-subtracted images to check for presence of a counter-image to further decide possibility of lensing.

Some stars are easier to spot than the others. For example, brighter stars cause saturation in the camera pixels (see bluish horizontal bars in the images below) and/or show diffraction spikes (bluish red rays). If you see any objects in the vicinity, these are most likely distant background galaxies in the line of sight.

The mass of an individual star is too small for it to produce multiple images of a distant galaxy. These stars, however, could act as a lens for other distant stars within our own galaxy, if they appear to cross each other from our perspective. This is called Microlensing. The HSC Survey project is not designed to find such events.

Go to the Top


Artifacts

The process of recording astronomical images into the nice maps of the sky you see here can lead to some interesting artifacts in the data that aren't real. In addition, there are real objects in the sky that are moving (either fast or slow) that can leave strange features in the data.

The images presented here are colour images made up of three channels: red, green and blue. Some of these artifacts only come from one of the channels, leading to pure red, green, or blue features. The actual astronomical filters we use in these colour channels are red: i-band (just outside the visible range that our eyes are sensitive to), green: r-band (middle wavelength, the reddest light our eyes can see), and blue: g-band (what our eyes would see as green). There are some examples of moving target artifacts below.

Fast-Moving Objects

These are examples of fast moving objects, such as satellites, asteroids, airplanes, etc. As the camera observes the sky for a fixed amount of time per exposure, fast-moving objects leave trails in the data. The object in the left panel was only in the field when the image was taken in the green filter, therefore it appears green. The object in the right panel is similar, but the image was taken in the red filter.

Slow-moving Objects

In the left image, you see several red spots in a line. This is a slow-moving object, such as a distant asteroid or a slow-moving satellite. The object appears multiple times because it has moved between exposures taken in the filters that make up this image. As the sky images are collected filter by filter (not necessarily on consecutive nights), moving objects like these get smeared out or appear as discrete images. Each image in each filter is further made up of several exposures that have been stacked and stitched onto each other. The right image is similar, except the object is present in the green filter.

Image Residuals

This image has data in only the redder wavelength. Hence, all the astronomical objects appear red. The blue and green channels has no actual image of the sky but there are some rectangular artifacts appearing most likely due to the image being at the edge of the chip.

Go to the Top


Fun activity

Build your own lens


You can create images similar to gravitational lenses using a wine glass. Try looking at a light source, like a small flashlight or a candle, through the base of the glass. Play around with the glass to produce arcs and rings.
Go to the Top