Preliminary Model of Saturated Flow Using Cellular Automata

9Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.
Get full text

Abstract

A fully-coupled from surface to groundwater hydrological model is being developed based on the Extended Cellular Automata formalism (XCA), which proves to be very suitable for high performance computing. In this paper, a preliminary module related to three-dimensional saturated flow in porous media is presented and implemented by using the OpenCAL parallel software library. This allows to exploit distributed systems with heterogeneous computational devices. The proposed model is evaluated in terms of both accuracy and precision of modeling results and computational performance, using single layered three-dimensional test cases at different resolutions (from local to regional scale), simulating pumping from one or more wells, river-groundwater interactions and varying soil hydraulic properties. Model accuracy is compared with analytic, when available, or numerical (MODFLOW 2005) solution, while the computational performance is evaluated using an Intel Xeon CPU socket. Overall, the XCA-based model proves to be accurate and, mainly, computationally very efficient thanks to the many options and tools available with the OpenCAL library.

Cite

CITATION STYLE

APA

De Rango, A., Furnari, L., Giordano, A., Senatore, A., D’Ambrosio, D., Straface, S., & Mendicino, G. (2020). Preliminary Model of Saturated Flow Using Cellular Automata. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11973 LNCS, pp. 256–268). Springer. https://doi.org/10.1007/978-3-030-39081-5_23

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free