The Brera Multi-Wavelet survey (BMW)
Astronomical observations can often tell us much more than what they were planned to study. In particular, when imaging a part of the sky to study one object, you obtain information about other objects at positions in the sky near the original science. This opens up the possibility of serendipitous science. Serendipitous observations have some obvious limitations. Because you do not get to choose the sightlines, you cannot plan beforehand which objects to observe, and so it is not very useful for studying a preselected list of objects. The great power of the serendipitous observations comes from their ability to tell you about the statistics, or populations, of objects. In effect, you are taking a random sampling of the sky (as long as you are careful about which observations you choose). So quantities such as the abundance or the spatial correlation of sources can be measured very accurately. This is particularly important for X-ray astronomy, because access to X-ray observatories has always lagged behind optical or radio astronomy (this is because X-rays are absorbed by the atmosphere, and so the observatory has to be in space). Past serendipitous X-ray studies have mostly used the PSPC detector on the ROSAT satellite. This is because the field of view of the PSPC and its sensitivity were the best available until the very recent advent of the Chandra and XMM-Newton missions. However, the majority of the observations taken by ROSAT were taken with another instrument, the HRI (High-Resolution Imager). The HRI has a lower sensitivity and a smaller field of view than the PSPC, so at first glance it would seem inferior for a serendipitous search. However, there are many more HRI fields available, and especially, there are many more extremely long observations in the HRI archive. That, and the superior angular resolution of the HRI, make it a fruitful source for serendipitous searches.