Why predictions fail

Symposium 2015 Rouyn-Noranda
Mr D. Kirk Nordstrom (US Geological Survey)
Predictions usually fail because of human arrogance and ignorance. We think we know things we don’t know and physical reality is more complex than our simple models can portray. We cannot succeed by disregarding human failures. Humidity-cell tests (HCTs) were designed to mimic weathering reactions in the laboratory for remediation planning. Few really questioned whether such tests were meaningful when applied to a large-scale mine waste site. What should be the optimal length of time to do HCTs and why? With highly variable test results, what applies to the field and why? Is the water flow in a column at weekly intervals relevant to water flow in the field? How does water and air flow in an HCT relate to water chemistry? Does the procedure consider microbial growth? In many situations, HCTs cannot be considered definitive or even helpful, yet they are widely used. Geochemical models have substantial limitations that are rarely identified when applied to mine sites. Does equilibrium apply or not? Do we know enough about rates of reactions to use models quantitatively? How well have models predicted anything that could be tested many years later? Models can be helpful but only for those who know their limitations and who understand hydrogeochemical processes.
Full Access to Technical Paper
PDF version for $20.00
Other papers from Symposium 2015 Rouyn-Noranda