We investigate the influence of local site effects on earthquake source parameter estimates using the LArge-n Seismic Survey in Oklahoma (LASSO). The LASSO array consisted of 1825 stations in a 25 km x 32 km region with extensive wastewater injection and recorded more than 1500 local events (M < 3) during spring 2016. We analyze the site amplification dependence on earthquake corner frequency (fc), seismic moment (M0), and stress drop estimated by modeling individual spectra. We evaluate and correct these site effects and compare the effectiveness of the correction to results using the spectral ratio method. We estimate local site amplification at each station using the average Peak-Ground-Velocity (PGV) of 14 regional earthquakes (~130 km away). The fc from the single spectrum method negatively correlates with site amplification, whereas M0 from the single spectrum method positively correlates with site amplification. This suggests the source parameters calculated by modeling individual spectra are biased by the local site effects. The high amplifications are typically located on young alluvial sedimentary deposits. We correct site effects by removing the trend between PGV and these two parameters in the regression analysis, which reduces the standard deviation of these parameters across the array and makes the calculated stress drop less site dependent. We compare corrections using other site-effect proxies such as the Root-Mean-Square (RMS) amplitude, surface geological formation, P-arrival-delay, and topographic slope. The PGV and the RMS corrections provide the greatest reduction of the spatial deviation of source parameters. In comparison, the spectral ratio method effectively removes the site effects using the Empirical Green’s Function (EGF) approach. The trends being removed by EGF are close to the apparent trends between the single spectrum estimated parameters and the PGV, which suggests the consistency of these different correction approaches. Our results provide a potential way to remove the site effects when only the main event spectrum is available and demonstrates the effectiveness of using the EGF approach for removing site effects. The resulting inter-station variability provides an estimate of the likely uncertainty in source parameters estimated from smaller numbers of stations.

Cross-correlation of fully diffuse wavefields averaged over time should converge to the Green’s function; however, the ambient seismic field in the real Earth is not fully diffuse, which interferes with that convergence. We apply blind signal separation to reduce the effect of spurious non-diffuse components on the cross-correlation tensor of the ambient seismic field. We describe the diffuse component as having uncorrelated neighboring frequencies and equal intensity at all azimuths, and an independent (i.e., statistically uncorrelated) non-diffuse component arising from a spatially isolated point source for which neighboring frequencies are correlated. Under the assumption of linear independence of the spurious non-diffuse wave outside the stationary phase zone and the constructive interference of noise waves within that zone, we can suppress the spurious non-diffuse component from the noise interferometry. Our numerical simulations show good separation of one spurious non-diffuse noise source component for either non-diffuse Rayleigh or Love waves. We apply this separation to the Rayleigh-wave component of the Green’s function for 136 cross-correlation pairs from 17 stations in Southern California. We perform beamforming over different frequency bands for the cross-correlations before and after the separation, and find that the reconstructed Rayleigh waves are more coherent. We also estimate the bias in Rayleigh wave phase velocity for each receiver pair due to the spurious non-diffuse contribution.