Advances in long-term soil-pollution monitoring of Switzerland


  • Based on a presentation given at EUROSOIL 2008, August 25–29, 2008 in Vienna, Austria.


The Swiss soil-monitoring network (NABO) was launched in 1984 and comprises currently 105 observation sites covering all characteristic land-use types across Switzerland. So far, the sampling periodicity was 5 y, and the fifth sampling campaign will be accomplished by end of 2009. The concentrations of Cd, Zn, Pb, Cu, Hg, Ni, Cr, Co, and F were measured. The major results and conclusions are: (1) Even topsoils in remote areas are to some extent polluted, mainly by Pb, Cu, and Cd. However, elevated concentrations can also be of natural origin, e.g., for F, Ni, Cr, and Cd. (2) Land use alone is often a rather unreliable indicator to discriminate soil pollution. (3) After the 2nd campaign positive, negative, or no temporal concentration changes were measured. From the 3rd campaign onwards, nonmonotonous (zigzag) evolutions were frequently observed. Therefore, no certified trends can be stated after three measurement campaigns during a period of 10 y. (4) Soil monitoring is an environmental time-series problem. The only way to detect reliable signals and trends earlier is to improve the overall measurement quality (precision and bias) and to shorten the measurement periodicity. (5) The causes of temporal soil concentration changes are complex and result from natural processes, anthropogenic processes, and methodological artifacts. Hence, not all soil concentration changes are due to anthropogenic inputs. Based on the state-of-art of our experience, some basic methodological requirements and recommendations can be supported for a “good soil-monitoring practice”: (1) Assurance of long-term continuity and consistency under changing boundary conditions as site conditions, methodologies, etc. (2) Implementation of a scientifically and politically appropriate spatial and temporal measurement resolution. (3) Long-term assessment of reliability (quality assurance) by adequate quantification of precision, bias, and confidence intervals along the whole measurement chain. (4) Documentation of all potentially relevant boundary conditions by suitable metadata. Only soil-monitoring results meeting these requirements are fit to support political decisions.