Communication of Uncertainty in Spatial Data to Policy Makers

There is increasing interest in the spatial aspects of many policy decisions. To be valid decision-malung tools it is important that maps not only convey what is known, but also where the uncertainties lie. This paper reports the results of tests conducted to assess the ability of various methods of error representation to accurately assist interpretation of maps. The work uses the predicted surface and associated error surface (both generated by a kriging function) to test the ability of prospective and current users to interpret the data presented, and to draw simple logical conclusions about it. The ultimate aim is to study the effect of user perception of accuracy on the real-life decisions made using interpolated spatial data.