A newly developed image analysis technique has significantly improved the sensitivity limits reached by the IBIS imager on board INTEGRAL, resulting in the deepest survey ever compiled of the entire sky in the energy range between 17 and 60 keV.

Pushing the instrument towards its very limits, the novel method discloses a vast number of previously undetected faint sources, galactic and extragalactic alike.

For more than seven years, the INTEGRAL observatory has been surveying the entire X-ray and gamma-ray sky and has accumulated a copious amount of exposure time, targeting both the crowded regions along the Galactic Plane and the high-latitude portions of the sky, this latter region being dominated by extragalactic sources.

Theoretically, a longer exposure time translates into an improvement in sensitivity, but this connection is not always straightforward: a number of systematic effects plague the observations and limit the sensitivity of the instruments despite the increased exposure time.

Hence, new techniques are sought, and implemented, in order to overcome these systematic effects and to fully exploit the instrument performance.

A successful example of this synergy is a novel image analysis algorithm, recently developed to improve the sensitivity achieved by IBIS, the Imager on Board the INTEGRAL Satellite.

Since high-energy photons, such as hard X-rays and gamma rays, cannot be focussed using traditional lenses and mirrors, IBIS is a coded-mask telescope, consisting of a metal plate (or mask) with a pattern of holes, which is placed on top of a detector. Photons coming from an astronomical source pass through the holes and, depending on their incoming direction, cast a series of shadows on the detector.

"The principle is very similar to that of a pinhole camera," explains INTEGRAL Project Scientist Chris Winkler. "From the pattern of dark and bright pixels, or shadowgram, recorded by the detector, it is possible to reconstruct the position in the sky and the intensity of the sources that produced the shadowgram. The reconstruction relies on complex image analysis techniques, which are unfortunately prone to a variety of systematic effects, especially along the Galactic Plane," adds Winkler.

The so-called Galactic Ridge emission, a strong, diffuse X-ray radiation coming from the Galactic Plane, represents a serious problem in this process, which is further complicated by the vast number of sources in the crowded field of the Galactic Centre.

"In order to isolate individual sources on the sky, images have to be cleaned by removing the background signal, which in turn has to be properly modelled," says Roman Krivonos, a researcher at the Max Planck Institute for Astrophysics and at the Space Research Institute of the Russian Academy of Science, who led the study.

In this case, the background includes emission from the Cosmic X-Ray Background, instrumental noise and, depending on the galactic latitude of the observed fields, additional Ridge emission from the Galactic Plane.

"The main source of systematic effects is the mismatch between the model of the background, used in the image analysis, and the true background, actually present in the images," Krivonos explains. "Our new method contains an improved model of the Ridge emission based on near-infrared observations, a good tracer of the galactic X-ray emission; this enables us to remove this particular source of systematic effects," he adds.

The algorithm also contains a further cleaning step, through which all large-scale artefacts, due to residual systematic effects and mimicking extended structures on the sky, are removed.

The newly developed method suppresses systematic effects almost completely in extragalactic, high-latitude fields, and yields a significant, albeit not total, removal also in the portion of the sky dominated by the Galaxy. Observations have now a more-or-less uniform sky background, enabling the detection of previously unnoticeable faint sources.

The result is the deepest all-sky survey compiled to date in hard X-rays, covering the energy range between 17 and 60 keV. The sensitivity has reached instrumental limits on extragalactic observations, where the IBIS imager aboard INTEGRAL is working at its maximum efficiency; on galactic fields, observations do not reach, but significantly approach, the instrumental limits, delivering a survey of the Galaxy with the best currently available sensitivity in this energy band.

"After 7 years of operations, IBIS has collected data over very long exposure times, and we can thus finally profit from the instrument's full capabilities. At this point in the mission's lifetime, such a technique, pushing the instrument towards its limits, is especially valuable," comments Winkler.

The catalogue of extragalactic sources detected in the study, mostly Active Galactic Nuclei (AGN), benefits enormously from the newly developed method, resulting in a much deeper survey than previously achieved. The higher sensitivity of the new observations enables a significant number of sources up to a redshift of z~0.1 to be detected, thus probing how the properties of hard X-ray emitting AGN evolved over the last thousand million years of cosmic history.

The sample of galactic sources, comprising compact sources of X-ray radiation such as accreting black holes and neutron stars, is also considerably larger than previous comparable catalogues. "It represents a prototype of the compact source samples that will be detected in nearby galaxies by future hard X-ray missions," says Krivonos.