The Lagoon Nebula
We here at Hubble Image Central (a.k.a. the Office of Public Outreach at the Space Telescope Science Institute), along with the entire Hubble Space Telescope team, love to celebrate the anniversary of our favorite orbiting telescope’s launch on April 24, 1990 by releasing a new picture. This year we’re happy to present a dramatic image of an amazing cosmic landscape called the Lagoon Nebula, also known as Messier 8 or M8. This image was made in early 2018 using Hubble’s primary imaging instrument, the Wide Field Planetary Camera 3 (WFC3).
What are we looking at?
Of course the Lagoon is not a tropical pool surrounded by a coral reef as the name suggests, but an immense cloud of gas and dust light-years across, actively churning out new stars. Along with other such places in our Milky Way Galaxy, it beautifully demonstrates the processes driving the formation of stars out of clouds of gas and dust, ultimately also making planets and stuff like us. The brightest, bluest stars formed not long ago – just a few million years at the most – an instant of cosmic time. These extremely hot stars like the one near the center of this small detail from the larger visible-light image produce copious amounts of extremely high energy light and a strong wind of fast-moving, energetic particles.
The wind and radiation carve out cavities in their “nursery” clouds and sculpt amazing, complex pillars and other structures. If there’s enough material left over, the wind and radiation from the first generation of stars compresses the remnants enough to form another generation of stars – rinse and repeat.
This time we have two images of the same region made by WFC3. One was made using the WFC3/UVIS camera that records visible and near-ultraviolet light, as in the above image. The other camera, WFC3/IR records near infrared light and produced a very different image (not precisely the same field but mostly overlaps the visible light image).
The first thing I noticed is how many more stars and how much less nebula we see. This is because the visible light from the stars is absorbed and scattered by the dust in the nebula but the infrared light passes through more easily. And the gas in the nebula is emitting a lot of light in the visible range but not so much in the infrared.
This is a great demonstration of the new science that will be available to the James Webb Space Telescope (JWST), NASA’s next generation flagship space science mission. JWST’s cameras and spectrographs will record infrared light, but in a wider range of wavelengths than Hubble, reaching into the mid-infrared (to about 25 microns).
Producing the visible light image
We used similar techniques as with many other Hubble images to produce the Lagoon Nebula images. (See an earlier blog post with more details about this process). We began with individual exposures from WFC3/UVIS in four filters. Because the camera sees only a very small region of sky (about 2.7 arcminutes square in a single frame) a small mosaic of 2×2 overlapping tiles covers the brightest part of the Lagoon. Here are the four images from one of the filters (F502N, sampling light emitted by ionized oxygen at 502 nanometers).
First we stitched these together to produce the mosaic for each filter:
Three of the filters, F502N, F656N, and F658N, sample very narrow bands of color just a few nanometers wide, transmitting the light of the elements oxygen, hydrogen, and nitrogen respectively. One of the side effects of these narrow-band filters is that starlight is reduced relative to light from the nebula gas because the stars emit light in a very broad range of wavelengths as opposed to the gas that emits light mostly in very specific wavelengths. The fourth filter, F547M samples a wider range of wavelengths (about 75nm wide centered on 547nm) so the stars appear brighter relative to the nebula. Exposures using this an similar filters are often used to further reduce the starlight to better analyze the nebula emission.
You may notice some artifacts that become more visible in the F547M filter. The broad, vertical streak at upper left is caused by light from a bright star scattering within the camera, even though the star itself is outside the view of the sensor. There are also a few “figure-8” features that are caused by light of bright stars reflected from the filter or other optical surfaces in the camera creating an out-of-focus and distorted image. These are quite subtle effects but appear when there are relatively bright stars in the field of view.
We apply color to the three narrow-band filters, using the three “additive primary” colors red, green, and blue, to the three filter images oxygen, hydrogen, and nitrogen respectively, and produce a preliminary composite.
Note that the applied color is not the actual color of the light that was recorded. In fact the wavelength of the light of hydrogen (656nm) and nitrogen (658nm) are so similar that their color appears identical. To extend the range of colors in the composite image, we apply green instead of red to the hydrogen image. Of course this means that the colors in the reconstructed image are not what our eyes would see. However, because we have separated two images from light of the same color, stretching out the color palette so we are visualizing more information from the data than if we applied the visible color.
In the final steps in post-processing, we adjusted the image to restore some of the contrast and structure that was recorded by the camera. The dynamic range in these images – the variation in light intensity in the recorded light – is enormous and challenging to manage.
You may notice that the stars in the image look somewhat unnatural with mostly the same oddly magenta color cast. That’s not because of their actual color but because of the nature of the filters and the colors we applied. Also, we were able to use exposures from another filter (F547M) that transmits a broader range of wavelengths to make the star colors more neutral and natural. Because the broader filter transmits relatively more light from the stars than from the nebula, the stars appear brighter. We adjusted the image to suppress the nebula light even more and added it in “Lighten” blend mode in the layering of the constructed image.
Producing the infrared image
The WFC3 instrument includes another camera that records infrared light. In this case we had images using two filters: sampling broad bands of wavelength around 1.25 and 1.6 microns. As with the visible light images, we applied different colors to the images. You might think since we have 3-color technology, we’d need to use all three but this works with just two, although it results in a limited color palette. We applied complementary colors red and cyan to the F160W and F125W filters respectively.
The rest of the post-processing is more subjective. I felt that shifting the colors produced a more pleasing palette and represents more realistic star colors. Since we have already translated the invisible infrared light into the visible range, the perception of color in the image is rather arbitrary.
As with the visible light image, we applied some brightness and contrast adjustments to bring out more details in the image, from darkest to brightest, resulting in a more interesting, dramatic result.
It’s always gratifying to be able to visualize an image from Hubble data and reveal details of the cosmos never seen before in such detail. Sometimes the images are especially dramatic and surprising, like these new images of the Lagoon. Watch this space for more background about this rich dataset and the process of visualization.