Much of modern astronomy has sprung from systematic surveys - the 3C, NVSS and FIRST radio catalogues, the Ariel V and Uhuru sky surveys, the IRAS far-infrared catalogue and the COBE sky-maps. Through much of this time the UK Schmidt photographic sky atlas (and its Palomar counterpart) has been the backbone of UK astronomy. As well as providing identifications of X-ray and radio sources, and saving huge amounts of 4-m telescope time by acting as a fundamental reference atlas, it has been a source of science in its own right (e.g. Abell clusters, Lynds dark clouds, the APM galaxy catalogue).
There are several arguments for the importance of large systematic surveys. (i) The first is the idea of a sky atlas as a fundamental resource, as support to other kinds of astronomy, as a leaping-off point by selecting samples of unnusual objects for detailed study, and as a gold-mine of hidden information whose future uses can only be guessed. (ii) Some science goals simply take large amounts of telescope time, for example because a very large sample is needed for statistical reasons. If a project takes more than say 100 nights, it is unlikely to be undertaken by individuals. Furthermore such ambitious goals will almost certainly be the aim of a whole community. A communal approach to collecting the data therefore makes sense. (iii) Sometimes an intrinsically large sky area is needed, as in studies of large-scale structure or the Milky Way. (iv) The same dataset can often be used for many different purposes - for example a large area of sky may be mapped to get a sample of 10 million galaxies, but somewhere in there are perhaps 10 cool halo white dwarfs that may be the solution to the dark matter problem. Once again, a communal approach is far more efficient.
Of course strong arguments can also be made for highly targeted research in small chunks of competitive open time. However with WFCAM there is an exciting opportunity for a truly ground-breaking survey project that will be of world significance. We can expect that the most productive and scientifically exciting results will flow from survey work whenever we move into a new region of parameter space. The growing importance of the IR is well recognised - for the detection of cool objects, for seeing through the extinction caused by dust, and as a shortcut to the high-redshift universe. The first IR sky surveys (DENIS and 2MASS) are just now taking place, but these are relatively shallow - more equivalent to, say, the Shapley Ames catalogue, than the Palomar/UK Schmidt surveys, let alone the SDSS. Our proposed survey programme will be the true match of the SDSS optical survey, but will also go well beyond it, as we probe deeper into smaller regions.