Avoiding Concentration Spikes when Simulating Cells that are Emptying (and Refilling)

 Posted by Rick Kossik

As part of the process of creating a very detailed online course for the GoldSim Contaminant Transport Module (for which a number of excerpts have been published in previous blog posts), I have been revisiting the most common questions and problems that we have been asked about by our users over the years in order to make sure to discuss these in the course (as well as to add more detailed descriptions, when warranted, in the User's Guide).

One of the more common issues that we see is the problem of GoldSim producing unrealistic spikes in computed concentrations when the volume of water (or another medium) in a Cell changes significantly over a timestep.  This most commonly happens when the Cell dries out (i.e., the volume actually goes to zero) and (perhaps) subsequently refills. There is a way to avoid this problem (which we have provided to users when the issue arises), but it is not discussed in detail in the current version of the User's Guide (which will be remedied in the next release).

In this blog post, I will first discuss the nature of the problem (when does it occur?) and how to address it in the current release of GoldSim (12.1).  For those that are interested, I will then discuss the numerical reasons for this problem, as well as an alternative approach that will be available for addressing the problem in the next GoldSim release (expected later this year).