Guidelines for validation and verification of groundwater mathematical studies (GULLIVER): a case study

Abstract

The goal of this work is to define a set of guidelines for “validating" mathematical models that are to be used in real-world problems of subsurface flow and contaminant transport. Groundwater models today are being used to study increasingly complex problems involving large temporal and spatial scales and with many uncertainties inherent in the data and in the models themselves. It becomes important therefore to establish procedures and benchmark tests with which to assess a given model's adequacy for simulating a specific groundwater problem. In the GULLIVER project we propose to create a public library of phenomenological schemes (PS), each one satisfactorily solved by an established model. A user with a specific application and model at hand can select the PS that most closely corresponds to his application, and corroborate his model against the established solution. The user's model is considered validated when the two numerical solutions match according to specified criteria. In this paper we describe this methodology and present its application to a 2D saturated flow and transport test case based on data from a field study site in Calabria, Italy.