Benchmarking the Sandbox: Quantitative Comparisons of Numerical and Analogue Models of Brittle Wedge Dynamics

BUITER, S.; SCHREURS, G.; ALBERTZ, M.; BEAUMONT, C.; BURBERRY, C.; CALLOT, J-P.; CAVOZZI, C.; CERCA, M.; CHEN, J-H.; CRISTALLINI, E.O.; CRUDEN, A.; CRUZ, L.; COOKE, M.; CROOK, T.; DANIEL, J.M.; EGHOLM, D.; ELLIS, S.; GERYA, T.; HODKINSON, L.; HOFMANN, F.; GARCÍA, V.; GOMES, C.; GRALL, C.; GUILLOT, Y.; GUZMÁN, C.G.; NUR HIDAYAH, T.; HILLEY, G.; KAUS, B.; KINKMÜLLER, M.; KOYI, H.; LANDRY, W.; LU, C-Y.; MACAULEY, J.; MAILLOT, B.,; MERIAUX, C.; MISHIN, Y.; NILFOUROUSHAN, F.; PAN, C-C.; PASCAL, C.; PILLOT, D.; PORTILLO, R.; ROSENAU, M.; SCHELLART, W.P.; SCHLISCHE, R.; SOULOUMIAC, P.; TAKE, A.; VENDEVILLE, B.; VETTORI, M.; VERGNAUD, M.; WANG, S-H.; WITHJACK, M.; YAGUPSKY, D.L.; YAMADA, Y.

San Francisco, Estados Unidos

Congreso; 2010 AGU Fall Meeting; 2010

American Geophysical Union

When numerical and analogue models are used to investigate the evolution of deformation processes in crust and lithosphere, they face specific challenges related to, among others, large contrasts in material properties, the heterogeneous character of continental lithosphere, the presence of a free surface, the occurrence of large deformations including viscous flow and offset on shear zones, and the observation that several deformation mechanisms may be active simultaneously. These pose specific demands on numerical software and laboratory models. By combining the two techniques, we can utilize the strengths of each individual method and test the model-independence of our results. We can perhaps even consider our findings to be more robust if we find similar-to-same results irrespective of the modeling method that was used. To assess the role of modeling method and to quantify the variability among models with identical setups, we have performed a direct comparison of results of 11 numerical codes and 15 analogue experiments. We present three experiments that describe shortening of brittle wedges and that resemble setups frequently used by especially analogue modelers. Our first experiment translates a non-accreting wedge with a stable surface slope. In agreement with critical wedge theory, all models maintain their surface slope and do not show internal deformation. This experiment serves as a reference that allows for testing against analytical solutions for taper angle, root-mean-square velocity and gravitational rate of work. The next two experiments investigate an unstable wedge, which deforms by inward translation of a mobile wall. The models accommodate shortening by formation of forward and backward shear zones. We compare surface slope, rate of dissipation of energy, root-mean-square velocity, and the location, dip angle and spacing of shear zones. All models show similar cross-sectional evolutions that demonstrate reproducibility to first order. However, we find differences in shear zone dip angle and surface slope between numerical and analogue models and, in 3D experiments, along-strike variations of structures in map view. Our experiments point out that we need careful treatment of material properties, discontinuities in boundary conditions, model building techniques, and boundary friction for sandbox-like setups. We show that to first order we successfully simulate sandbox-style brittle behavior using different numerical modeling techniques and that we can obtain similar styles of deformation behavior in numerical and laboratory experiments at similar levels of variability.