Automated Identification of Sedimentary Basins in California and Nevada Using Wavelet Decomposition of Topography

Abstract Accurate identification of sedimentary basins and modeling of their amplification effects are essential for seismic hazard analysis, particularly in regions where ground motions are strongly influenced by basin structures. Traditional basin identification methods often rely on manual interpretation, which is limited by data coverage and by uncertainty from subsurface velocity models or subjective judgment. We present an automated method for identifying sedimentary basins using digital elevation models (DEMs) and estimating their horizontal extent. Topography often reflects subsurface structure, with basins typically appearing as low, flat terrains formed by sediment accumulation in natural depressions. We apply the continuous wavelet transform to decompose topography into multiple length scales. Sedimentary basins exhibit low-amplitude coefficients at short scales due to minimal surface roughness, and negative coefficients at larger scales corresponding to their lateral extent. Our automated approach identifies basins across California and Nevada, producing results consistent with traditionally mapped basins in California and revealing basin areas in Nevada. This technique provides spatially continuous and reliable basin identification and estimates of basin size, which serves as a horizontal metric independent of traditional depth metrics such as Z1.0 and Z2.5. This approach can be readily applied to sedimentary basins in other regions worldwide that exhibit clear topographic expression, given the broad availability of DEMs. These basin terms can be used as predictors in ground-motion models to include basin effects.