An 850μm SCUBA map of the Groth Strip and reliable source extraction

We present an 850μm map and list of candidate sources in a sub-area of the Groth Strip observed using SCUBA. The map consists of a long strip of adjoining jigglemaps covering the southwestern 70 arcmin of the original WFPC2 Groth Strip to an average 1 σ rms noise level of ≃ 3.5mJy. We initially detect 7 candidate sources with signal-to-noise ratios (SNRs) between 3.0 and 3.5 σ and 4 candidate sources with SNR ≥ 3.5. Simulations suggest that on average in a map this size one expects 1.6 false positive sources ≥ 3.5 σ and 4.5 between 3 and 3.5 σ. Flux boosting in maps is a well known effect and we have developed a simple Bayesian prescription for estimating the unboosted flux distribution and used this method to determine the best flux estimates of our sources. This method is easily adapted for any other modest signal-to-noise survey in which there is prior knowledge of the source counts. We performed follow-up photometry in an attempt to confirm or reject 5 of our source candidates. We failed to significantly re-detect 3 of the 5 sources in the noisiest regions of the map, suggesting that they are either spurious or have true fluxes close to the noise level. However, we did confirm the reality of 2 of the SCUBA sources, although at lower flux levels than suggested in the map. Not surprisingly, we find that the photometry results are consistent with and confirm the de-boosted map fluxes. Our final candidate source list contains 3 sources, including the 2 confirmed detections and 1 further candidate source with SNR > 3.5 σ which has a reasonable chance of being real. We performed correlations and found evidence of positive flux at the positions of XMM-Newton Xray sources. The 95 per cent lower limit for the average flux density of these X-ray sources is 0.8mJy.