Site-specific management in agriculture and of contaminated sites requires accurate maps of the soil properties of interest. Viscarra-Rossel and McBratney (1998) indicated that contour maps of such properties are best produced by grid sampling followed by geostatistical analysis. The latter, usually involves computing and modelling the variogram followed by kriging with the data and the appropriate model parameters. The most usual method of estimating the variogram is by the method of moments. Webster and Oliver (1992) recommended 100 to 150 sampling sites to estimate the MoM variogram accurately. A sample of this size per field or site is often beyond the budget for surveys of many farms and contaminated sites. A common approach to agricultural sampling is one sample per hectare, which even for a large field of 50 ha would not provide an adequate sample size to compute the MoM variogram. Furthermore, this approach takes no account of the spatial scale of variation present. Kerry and Oliver (2003) showed that variograms from ancillary data could be used to guide sampling. If the variation shows strong continuity, the variogram range might be large in relation to the size of the field and the resulting sample size would be too small to compute a reliable MoM variogram. A sample size of 100 – 150 sites, however, would result in a considerable waste of sampling effort. For many reasons, therefore, there is a need for alternative methods to compute variograms reliably with fewer soil data. A possible solution is to use standardized variograms computed from aerial photographs and ECa data to krige standardized soil data that have been sampled at an appropriate interval but are still sparse. Intensive data for soil clay content from four field sites of different size, physiography and parent material in southern England were sub-sampled to produce various sampling schemes. The data were then used for cross-validation using the model parameters from variograms of the soil properties, standardized variograms from ancillary data and variograms from ancillary data with the nugget and sill variances set using close and widely spaced data points from a targeted sampling scheme based on bare soil reflectances. The mean squared deviation ratios (MSDRs) and median squared deviation ratios (MeSDRs) from cross-validation with each sampling scheme and type of variogram will be presented. Maps produced by the various variograms using sparse, yet appropriately spaced soil data will also be compared with those produced by calculating and kriging with the variogram from intensive soil data. References: Kerry, R. and Oliver M. A., 2003. Variograms of ancillary data to aid sampling for soil surveys. Precision Agriculture 4, 261-278. Viscarra-Rossel, R.A. and McBratney, A.B., 1998. Soil chemical analytical accuracy and costs: implications from precision agriculture. Australian Journal of Experimental Agriculture 38, 765-775. Webster, R. and Oliver, M.A., 1992. Sample adequately to estimate variograms of soil properties. The Journal of Soil Science 43, 177-192.