Data extrapolation to real situations is difficult and this will be dependent on several parameters (e.g.):

• Location of the tests (laboratory or field conditions, to see whether the environmental conditions of the site were replicated)
• Type, kinds and morphology of wastes.
• Amount of wastes in the cells. Is the amount of waste used in the cell good enough to replicate the behavior of wastes that will be deposited in the WRD or TSFs, etc.?
• Physical properties of wastes and hydraulic conditions in the cells/columns, etc.

If you indicate your purpose and other objectives, this can be further discussed.

This situation is very complex but, if you can use PHREEQCI (Inverse Mass Balance Model) you can calculate weekly mineral solubility using physical and chemical data from water analyses obtained.

It is possible certainly with a regression analysis or something but that only tells you about what might happen in the future of the humidity cell test not in field conditions so it depends what you want to get out of it. If it’s for field estimation of rates then humidity cells are in most cases not that useful. Field parameters that matter include particle size, gas flux, liquid: solid ratio, temperature, physical competence of the material. Many people have published scale up factors for these tests but the results are in many cases a bit of a scatter gun with very little correlation if looked at objectively. Personally I prefer to run with newer and more advanced kinetic tests that can simulate field conditions rather than reverse engineer something from the traditional analysis. Just my opinion and I sure there are plenty of people waiting to defend the humidity cells but technology has progressed somewhat in this area now so there are better alternatives.

It depends on what you conclusions you are trying to extrapolate. SRK uses HCTs extensively to predict drainage characteristics of waste rock, tailings and reactive surfaces. The trick is the use of scaling factors to modify the HCT results. The scaling factors should account for many of the modifying factors that Steve lists. It is always a good idea to calibrate the scaling factors using bin or drum tests, using large volumes of the waste rocks under actual field conditions. Another issue is that HCTs usually do not run long enough to simulate the time that the waste material will be exposed. However, you can get a general idea of the depletion rates of AP and NP using the HCT, mineralogy and static data.