Small-scale lithospheric heterogeneity characterization using Bayesian inference and energy flux models

SUMMARY Observations from different disciplines have shown that our planet is highly heterogeneous at multiple scale lengths. Still, many seismological Earth models tend not to include any small-scale heterogeneity or lateral velocity variations, which can affect measurements and predictions based on these homogeneous models. In this study, we describe the lithospheric isotropic structure in terms of intrinsic, diffusion scattering quality factors, as well an autocorrelation function, associated with a characteristic length (a) RMS fractional fluctuations (ε). To obtain characterization, combined single-layer multilayer energy flux new Bayesian inference algorithm. Our synthetic tests show technique successfully retrieve input parameter values for 1- 2-layer algorithm resolve whether data be fitted by single set parameters range required instead, even very complex posterior probability distributions. We applied three seismic arrays Australia: Alice Springs array (ASAR), Warramunga Array (WRA) Pilbara Seismic (PSAR). model results suggest intrinsic attenuation are strongest ASAR, while total similarly strong ASAR WRA. All factors take higher PSAR than other two arrays, implying beneath less attenuating The crust more mantle all arrays. Crustal correlation lengths ∼0.2 1.5 km ∼2.3 3.9 per cent, respectively. Parameter upper unique, combinations low (a < 2 ε ∼2.5 cent) being likely those high variations > 5 respectively). attribute similarities WRA their location proterozoic North Australian Craton, opposed PSAR, lies archaean West Craton. Differences ascribed tectonic histories regions same craton. Overall, highlight suitability combination future studies, since approach allows us compare also giving detailed information about trade-offs uncertainties determination parameters.