Jatnieks, JanisSips, MikeLucia, Marco DeDransch, DorisA. Middel and K. Rink and G. H. Weber2015-05-242015-05-242015https://doi.org/10.2312/envirvis.20151088Geochemical models serve wide-ranging geoscientific applications for underground resource exploitation, aquifer remediation, gas storage and similar problems. Such models are time consuming to calculate. Replacing the fullcapability simulation model, for each element of spatial discretization, with a fast surrogate is a promising acceleration approach for this problem. The balancing of speed and accuracy trade-off inherent to the surrogate modeling approach requires expert involvement and is best supported interactively. In this paper we argue that Visual Analysis has a prominent role to play for facilitating this process. It allows to involve expert knowledge regarding the specific characteristics of application scenario as part of the surrogate creation process. We describe the core problem of accelerating geochemical simulation by surrogate models and outline a strategy for approaching it with Visual Analysis.J.2.8 [Computer Applications]Physical Sciences and Engineering—Earth and Atmospheric SciencesI.6.5 [Simulation and Modeling]Model Development—Modeling methodologies10.2312/envirvis.2015108831-35