Soil lead distribution at an abandoned smelter site in Socorro, New Mexico

Soil Pb concentrations from three 30.5-m× 30.5-m squares at an abandoned smelter site in Socorro, New Mexico, are reproducible only within several hundred ppm. There was no statistically significant difference in log-transformed sample means for two of the three squares tested, due to the large degree of variability among samples from each square. Even where sample means are not significantly different, attempts at replication using similar sample spacing and collection techniques yield radically different spatial distributions of Pb. This variability suggests that common assumptions about soil contamination at hazardous waste sites must be reconsidered and also that understanding the statistics of contaminant distribution may be the key to designing effective clean-up strategies.