This dataset contains diffusion-sorption data, generated with numerical simulation based on three different sorption isotherms, namely the linear, Freundlich, and Langmuir isotherms. This dataset is used to train, validate, and test all the deep learning models that are used in the publication "Learning Groundwater Contaminant Diffusion-Sorption Processes with a Finite Volume Neural Network". The dataset for each sorption isotherm includes the dissolved and total contaminant concentration data, as well as spatial coordinates and timestamps that correspond to the concentration data.

More detailed information is also provided in our Github repository (https://github.com/CognitiveModeling/finn) and our submitted paper to the Water Resources Research journal.