Jan 26

Time, 2nd part


When you want your work to survive over time, one solution is to remove time from the equation.

One good example is the Calar Alto image we are finishing now. We have been imaging the Stephan’s Quintet compact group of galaxies during the last four years. We acquired a total exposure time of 59 hours. During 2009, 2010 and 2011 we used broadband RGB filters with the old, 4-megapixel camera. Last year, with the new 16-megapixel camera, we acquired 15 hours of high-resolution, unfiltered images; these exposures were made in very steady nights so the FWHM value of the combined image is ~0.9″. Complementing the wideband images we acquired also narrowband, H-alpha images to record the effects of the interaction between the galaxies.

The use of this filter set let us to build a very interesting photo, in my opinion. It will comprise six different wave bands. First we have the usual RGB wideband color information:


From the unfiltered image (which contains not only visible light, but also utraviolet and infrared), we can subtract the RGB images in order to isolate the light outside the visible spectrum (mainly infrared light). These isolated sources can be used to enhanced the IR-emitting sources by adding it to the color image. Below we can see these sources as tiny orange-colored dots all around:


Finally, we have two H-alpha lines. The first is at its normal wavelenght of 653 nm, and shows the emission nebulae of one of the galaxies, which is much closer to us than the other ones:


We are going to show this emission as pink:


Finally, to record the Hydrogen alpha emission of the other galaxies (which have a radial velocity of 6,600 km/s) we had to use a narrowband filter centered at 667 nm. This filter shows very well the shock wave between the galaxies:


This emission will show in our palette as red. Summarizing, we have an image with visible-light RGB colors, enhanced IR sources, and two different hydrogen emission lines:


In astrophotography there are many concerns to achieve a good work; time shouldn’t be one of them. In this type of works, it’s not easy to convince yourself if it’s worth the time. Maybe the correct question is “it’s worth the image?” Looking at the results above, for this image I would say definitely “yes!”

Jan 12

The nature of light, 1st part

This is a very important topic to me. In nature photography, communication with the public —from a documentary point of view— can be achieved only if there is a close connection between the image and nature.

In astrophotography, this intimate connection is contained in what we call the “master image”. This is the state where the image is closest to represent the nature of the light emitted by the photographed object. This image has some characteristic facts, the most important being its linearity: the light flux emitted by the objects is proportional to its corresponding pixel value. In this state, our eye cannot see efficiently the image because we have a non-linear response to light flux. This can be better depicted in the below image:


A linear image has usually too much contrast, resulting in an extremely dark view. Fortunately, we have the right tools to “translate” the image to the brightness shades that our eyes use to see, as the right image shows. The main fact about the linear image is that it opens the possibility to correct instrumental effects on the image. In these effects we may include, among others, vignetting, thermal noise, dust shadows or light pollution gradients. In other words, our linear data give us the opportunity to reach an image that represents exclusively the natural fact that we’re photographing.

During the last years I have been developing new techniques to achieve a higher fidelity of my master images from Calar Alto. One point that is extremely important is the correction of artificial illumination gradients. A good example is this image of the spiral galaxy Messier 63, acquired at the Astronomical Observatory of Aras de los Olmos:


This image shows clearly a diagonal light gradient coming from the bottom right corner which doesn’t correspond to any object in the sky. This gradient is actually limiting the visibility of faint objects in our photo, as well as the fidelity of the scene. The only photo we can generate from these data is one showing the main body of the galaxy:


The technique to remove the image gradients is based on a supplementary image set that we call “super flat frames”. This images are acquired in between the exposures of our object by pointing the telescope at an offset of the object field. In the field of view of the super flat frames we’ll have only stars and no extended objects. The goal of these frames is to reproduce the light gradients present in our object exposures —but without it, of course.

By photographing this adjacent field, we can generate a model of these gradients:


And this image can be used to remove the gradients from the image of our object. Although this technique requires additional work time at the telescope, we are going to reach much more deepness in our image, as show the below result:


This deeper image shows that the galaxy is actually surrounded by a faint, annular-like structure. This is a “dead” galaxy that is being to be absorbed by the main galaxy. It is extremely faint and diffuse, so it requires a very clean sky background. To better look at these faint structures, below you have the inverted image with some noise reduction:


Gradient correction is even more important with color imaging. Here you have an example with the small galaxy group HCG92, acquired from Calar Alto Observatory. This is the image prior to gradient correction:


And after correction:


The most important concept to understand about these techniques is the connection they reveal between data acquisition and image processing. A more comprehensive article is on the way, just stay tuned to this blog and to PixInsight website.

Jan 09



Life at an astronomical observatory can be summarized in three words: astronomy, gastronomy and hiking.

When one is going to be completely isolated on the mountain for one or two weeks, organization is very important. Here on the mountain, it involves not only our astro-related work -which is very time consuming, specially in winter-, but also other activities I consider to be essential to not become an observatory-maniac.

One of these activities is gastronomy. Some months ago I bought a cooking book, “The family meal: home cooking with Ferràn Adrià” (the Spanish version is titled as “La comida de la familia”), written by Ferràn Adrià and the staff of the El Bulli restaurant. It is not a good book; it is a genius book, and what it was supposed to be in my opinion. The book has a fantastic introduction, followed by a collection of three-course menus. The goal of the book is to move the restaurant methodology to a small-scale, every-day home eating.

Don’t be scared by the author, this is a book about traditional cooking with easy to find ingredients. What I really like is that it is written in the same manner as I write my own image processing articles. The text doesn’t show explicitly every single detail, so there is a small room for the reader to experiment, to improvise, to imagine. This leads from a mere recipe to a text that can be used as a basis for your art, whether we’re speaking of cooking, photography or painting.

These are some chocolate truffles (above photo). Instead of adding some alcohol I’ve made a little experiment by melting balsamic candy and honey with the cream used to melt the chocolate. It is recommendable to use a good chocolate. I used Valrhona; it’s not cheap, but worth the try.

Jan 03

Time, 1st part

Every image should have a unique value. This simple idea is a good ingredient for a successful image over time. This is a question I always face with my photos, and is precondition for each of my works to be published.

After several weeks of work with LBN552, I just realized which are the unique values of this image. Is as simple as looking at the two main subjects of the photo: stars and dust.


The field of LBN552 is not plenty of fine detail with other objects like M51 or the inner region of M31. This means that the stars will have in this image a prominent role that they don’t use to have in other images. Several years ago, when I was starting in astrophotography, stars used to be flat, white disks with almost any color information, and sometimes surrounded by a dark ring. The problem, at that time, wasn’t the techniques nor the aesthetics. The problem is that there really wasn’t any aesthetic point of view. Image processing was understood as a set of techniques used for correcting problems and enhancing the image.

Today we understand image processing in astrophotography as the set of techniques used to communicate the image, making it an aesthetic language itself. That’s why dynamic range compression and color preservation is so important in stars. Assuming that each structure in the image, by its single characteristics, needs a different processing, we can achieve an image where each object has its unique value. In the case of stars, this means color, which will show up as a specific vibration if we avoid their saturation.

The techniques I’m applying to this image are derived from the ones described in my processing tutorial “Dynamic range and local contrast”, based on an image of NGC7023.


The second unique value of this image is dust. There were to initial ideas to calibrate the color of this image. The first was to take as white balance a face-on spiral galaxy. This produced a bronwish / yellowish nebula, which is good for a subject like this (a molecular cloud). The second idea was to enhance the local color variations inside the clouds, which is always very interesting. This is attained by taking as white reference the object itself.

Both results have their respective trade-off. Conceptually they are correct, but the first shows only that the nebula is, in fact, brownish. And the second enhances too much the bluish parts of the clouds, so there is the risk of not making understandable -from a visual point of view- that this is a molecular (interstellar dust) cloud. The solution has been very simply: make an average of both color-calibrated images.

Color through stars and dust. This is the unique value of this image.

Dec 29

A Year Ending

It always happens when the year is about to finish. Suddenly, a conversation with Juan turns around our expectations and projects for the next year. And we both get excited when we see the amount of work we have to do and all the things that are coming in the near future. Maybe it’s an advantage we have due to our position, in the sense that we can have a global vision of how things are evolving in projects like PixInsight. But, anyway, in my opinion, it’s a very good practice to  make a reflection on these days about how things are going to evolve in the next year. Also about what we are going to do, and about what we have done in the present year as well.

2013 is going to be promising for PixInsight. Version 1.8 is about to be released in the next year, but what looks promising is not the new and future versions but how community is evolving around the platform. Juan’s bet about building PixInsight not only as a software package but as a whole platform is giving now its real results. Just a look to the officially published user scripts gives an idea of what I’m talking about. Next year we’ll see new interesting tools from PixInsight users. For instance, we’ll be able to start acquiring data from PixInsight and we’ll see a photometry script as well. Further automation tools will be developed, not only for image processing but also for observatory-related tasks. At large term, PixInsight will be used in professional observatories (where it is starting to be used), where it will work for data pipeline management and even for outreach with the use of automated processing sequences. All in all, in the next years PixInsight will have a much more extended use, inside and outside (where I have successful experience) the astronomy world.

2012 hasn’t been for me a great astrophotography year, as we have only released one Calar Alto image, but I have done an important contribution in forensics in collaboration with the National Police Dept. Next year we will continue to have about 50 observing nights at Calar Alto and we’re finishing two new images. Moreover, two new projects may happen in the near future. Either of them, if become reality, could be much more revolutionary that our work at Calar Alto.

Stay tuned.

Dec 18

Calar Alto Observatory

Zeiss123m In 2009, Calar Alto Observatory started an astrophotography project in collaboration with the Descubre Foundation. The goal of this project was to create a invaluable photographic archive of astronomical objects. This work was ordered to the Documentary School of Astrophotography and, from then, we have had the opportunity to work with professional instrumentation for 200 nights.

I want to remark the invaluable value of this project because it has unique aspects in the world. The first reason is the team. We’re not astronomers. We’re not photographers. We’re astrophotographers. And the astrophotography, as we do understand it, comes up from a dialog between art and science.


At this point, you may be thinking that I’m saying quite evident things. But that’s not the case in astrophotography (and astronomy) world. Until now, all the professional astrophotography projects have been relying on professional astronomers assuming that you only need a larger, better and sharper telescope to produce “spectacular” images. Kinda Hollywood special effects dept. Our project shakes the concept of interdisciplinary as we understand it today. Science divulgation through photography needs people with a good artistic basis and an understanding of what’s happening in the photo.

Another key point of this project is dedication. The fact of keeping focused in the same object over a large number of nights allow to capture the scene with a reliability and a deepness never known until today for an instrument of this size (a telescope with a 1.23 meter mirror). Some of our photos have the same or better deepness and definition of images acquired with some of the world larger telescopes, and also have a wider field of view and better color information.

And the third reason why this project is unique is the fact that it has been ordered to the Documentary School of Astrophotography. We are the first school of thought in the astrophotography world. As our founding statement says, the members of this school, “conscious of the critical moment that this discipline is facing at present, want to contribute our point of view and ideas to help define what astrophotography is and what it should not be. [...] We believe that astrophotography must preserve its documentary nature, as it has been doing historically. Along with providing transcendental scientific results, the documentary nature of astrophotography is fully compatible with the artistic vision that most astrophotographers have.”

“Astrophotography is not compatible, in our opinion, with image manipulations leading to generation of false information that is not contained in the original data. Some may think that such manipulations form part of a particular artistic vision, but we believe that they cannot be accepted as astrophotography. We consider that it is absolutely necessary to stay away from the idea, more and more widely accepted, that any procedure or technique can be accepted as valid in this discipline.”

Up to now, no other -professional or amateur- astrophotographer has started a similar initiative. Therefore the works we are publishing within this project have the added value of being the only astrophotos in the world where the authors have publicly signed an statement in which they get such a commitment.

You can read some of my tutorials and processing examples, as well as some video tutorials, at PixInsight website. They describe specific techniques I use to apply to our images. More are coming in the next years, but my intention in this blog is to speak about the aesthetic basis and the physics involved in our works in order to help you to better understand them. Keep tuned.

Dec 12

Astrophotography Notes


Welcome to my blog. Let me introduce myself. I’m an astrophotographer working at the Astronomical Observatory of the University of Valencia, in Spain. If you entered in this website, perhaps already know that I’m leading an astrophotography project in collaboration with Calar Alto Observatory (in southern Spain).

I need to share many things related with this project and my actual experience with the astrophotography art, thus I created this blog. Here you’ll read about general art, astrophotography, and a bit of science, always from my particular point of view.

Next post is coming very shortly. I’ll speak about the general aspects of the astrophotography project at Calar Alto. I hope these readings will help you to better look at my works, because the art in photography starts just at this point: when you’re looking at.