Numerical evaluation of uncertainty propagation in transport models:
non-multiGaussian heterogeneous formations

Gerónimo Llerar, Daniel Fernàndez-Garcia, Eduardo F. Cassiraga, J.Jaime Gómez-Hernández.

Institute for Water and Environmental Engineering, Polytechnic University of Valencia, Valencia, Spain,
gelleme@doctor.upv.es, dafernan@dihma.upv.es, efc@dihma.upv.es, jaime@dihma.upv.es

ABSTRACT

This paper evaluates the propagation of uncertainty in solute transport models through upscaling. Extensive Monte Carlo simulations were conducted at different computational support scales. Attention is paid to the difference in the propagation behavior of uncertainty between multi-Gaussian heterogeneous models that reflect high connectivity in the mean values and non-multi-Gaussian heterogeneous models that consider high/low connectivity of extreme values. The results demonstrate, that the loss of smallscale information by upscaling aquifer properties to construct a numerical model largely modifies the true heterogeneous structure of the aquifer compromising the final predictions of solute transport.