The brera multi-scale wavelet HRI cluster survey I. Selection of the sample and number counts

We describe the construction of the Brera Multi-scale Wavelet (BMW) HRI Cluster Survey, a deep sample of serendipitous X-ray selected clusters of galaxies based on the ROSAT HRI archive. This is the first cluster catalog exploiting the high angular resolution of this instrument. Cluster candidates are selected on the basis of their X-ray extension only, a parameter which is well measured by the BMW wavelet detection algorithm. The survey includes 154 candidates over a total solid angle of similar to160 deg(2) at 10(-12) erg s(-1) cm(-2) and similar to80 deg(2) at 1.8 x 10(-13) erg s-1 cm(-2). At the same time, a fairly good sky coverage in the faintest flux bins (3-5 x 10(-14) erg s(-1) cm(-2)) gives this survey the capability of detecting a few clusters with z similar to 1-1.2, depending on evolution. We present the results of extensive Monte Carlo simulations, providing a complete statistical characterization of the survey selection function and contamination level. We also present a new estimate of the surface density of clusters of galaxies down to a flux of 3 x 10(-14) erg s(-1) cm(-2), which is consistent with previous measurements from PSPC-based samples. Several clusters with redshifts up to z = 0.92 have already been confirmed, either by cross-correlation with existing PSPC surveys or from early results of an ongoing follow-up campaign. Overall, these results indicate that the excellent HRI PSF (5" FWHM on axis) more than compensates for the negative effect of the higher instrumental background on the detection of high-redshift clusters. In addition, it allows us to detect compact clusters that could be lost at lower resolution, thus potentially providing an important new insight into cluster evolution.