WFCAM is under construction at the United Kingdom Astronomy Technology Centre (UKATC) in Edinburgh, and is scheduled to be delivered to Hawaii in September 2003. The project scientist is Mark Casali, and the Project Manager is Dave Lunney. The work is undertaken in collaboration with the Joint Astronomy Centre (JAC), who operate UKIRT, and in part through an MOU with Japan, who are providing some of the detector arrays. WFCAM should start regular science operations at the beginning of 2004. The project website is at http://www.roe.ac.uk/atc/projects/wfcam, and a summary of current status is in a recent SPIE paper.
WFCAM has the widest field of view of any IR camera so far attempted, and being placed on a 4m class telescope, is fast enough to make to large deep IR surveys that are a scientific match to optical surveys such as the Sloan Digital Sky Survey (SDSS). It is based on a novel optical design using a quasi-Schmidt camera placed forward of Cassegrain focus that gives a 1 diameter field of view. It will be used in conjunction with UKIRT's tip-tilt secondary, which primarily corrects for wind-shake and dome seeing and is expected to deliver median seeing of 0.4 across the whole of this field of view. The camera employs four Rockwell PACE 2048 HgCdTe arrays operating across the JHK spectral region, with 0.4 pixels, and placed at 90% spacing, giving an instantaneous coverage of 0.19 sq.deg. Figure 1 shows the footprint on the sky compared with other IR cameras, which spells out what a radical advance WFCAM is, especially bearing in mind that the 2MASS detectors are on a telescope with an order of magnitude less collecting area than UKIRT.
The pixel size of WFCAM is somewhat undersampled with respect to the expected median seeing, but micro-stepping, as employed successfully with 2MASS, can produce well-sampled images. It will be possible to dither/mosaic WFCAM in a variety of ways, but a standard filled-in ``tile'' can be made by making 22 micro-steps and 22 macro-steps, resulting in a final image with 0.2 pixels and diameter 0.87, including small overlap regions that are observed more than once. In the K-band, exposures will not be more than 10 sec in order to avoid sky saturation, and this is likely to be taken as a standard stare time. Very little time is lost between micro-steps, but the macro-steps use telescope movement and involve a finite time to settle. With a standard 10 sec exposure, we expect the net observing efficiency to be 65%. The complete filled-in tile with sixteen separate exposures therefore takes approximately four minutes. Given the well known UKIRT performance and the expected throughput of WFCAM, then for point sources and an assumed seeing of 0.4 , and to a sensitivity of 5, we expect to reach a depth of K=18.4 in such a four minute observation - fifty times deeper than the 2MASS survey. In a ten hour night, and allowing two hours for calibration, one could survey 115 square degrees.