Comparison of observed ground-motion attenuation for the 2016/04/16 Mw7.8 Ecuador megathrust earthquake and its two largest aftershocks with existing ground-motion prediction equations

A megathrust subduction earthquake (Mw7.8) struck the coast of Ecuador on April 16th, 2016 at 23h58 UTC. This earthquake is one of the best-recorded megathrust events up to date. Besides the mainshock, two large aftershocks have been recorded on May 18th, 2016, at 7h57 (Mw 6.7) and 16h46 (Mw6.9). These data make a significant contribution for understanding the attenuation of ground motions in Ecuador. Peak ground accelerations and spectral accelerations are compared with four ground-motion prediction equations developed for interface earthquakes, the global Abrahamson et al. (2016) model, the Japanese equations Zhao et al. (2006) and Ghofrani and Atkinson (2014), and one Chilean equation Montalva et al. (2016). The four tested GMPEs are providing rather close predictions for the mainshock at distances up to 200km. However, our results show that high-frequency attenuation is greater for backarc sites, thus Zetal2016 and Metal2016, which are not taking into account this difference, are not considered further. Residual analyses show that G&A14 and Aetal2016 are well predicting the attenuation of ground motions for the mainshock. Comparisons of aftershock observations with Aetal2016 predictions indicate that the GMPE provide reasonable fit to the attenuation rates observed. The event terms of the Mw6.7 and Mw6.9 events are positive but within the expected scatter from worldwide similar earthquakes. The intra-event standard deviations are higher than the intraevent variability of the model, which is partly related to the poorly constrained VS30 proxys.

💡 Research Summary

The 2016 April 16 Mw 7.8 megathrust earthquake that ruptured the Ecuadorian subduction zone, together with its two largest aftershocks (Mw 6.7 and Mw 6.9 on 18 May 2016), provides one of the most complete ground‑motion datasets for a subduction interface event to date. This study exploits the dense network of permanent and temporary accelerometers that recorded peak ground accelerations (PGA) and 0.01–10 Hz spectral accelerations (SA) for these three events, and evaluates how well four published ground‑motion prediction equations (GMPEs) reproduce the observed attenuation. The GMPEs examined are: the global Abrahamson et al. (2016) model, the Japanese Zhao et al. (2006) formulation, the Ghofrani & Atkinson (2014) model developed for the Australia‑Pacific region, and the Chilean Montalva et al. (2016) equation. All four are interface‑type models that incorporate magnitude, distance, depth, and a site term based on VS30.

The analysis first separates the recordings into trench (fore‑arc) and back‑arc stations, because previous work has shown that high‑frequency energy decays more rapidly in the back‑arc. For distances up to 200 km the four GMPEs predict PGA and SA within roughly ±15 % of the observations for the mainshock. However, at back‑arc sites the high‑frequency (≥2 Hz) attenuation is noticeably stronger than the models anticipate. Consequently, the Zhao et al. 2006 and Montalva et al. 2016 equations, which do not distinguish between trench and back‑arc propagation paths, systematically over‑predict the observed SA by 30–50 % at frequencies above 5 Hz. By contrast, Ghofrani & Atkinson 2014 and Abrahamson et al. 2016 incorporate either separate attenuation coefficients or a non‑linear VS30 amplification that partially captures the back‑arc effect, resulting in smaller residuals.

Residual analyses were performed on a logarithmic scale. For the mainshock, the event terms (mean residuals) of G & A 2014 and Abrahamson et al. 2016 are essentially zero (‑0.02 to +0.03), and their intra‑event standard deviations (σ_intra) of 0.22–0.24 match the within‑model sigma (≈0.20–0.25). The Zhao et al. 2006 and Montalva et al. 2016 models show positive event terms (+0.10 to +0.15) reflecting the high‑frequency over‑prediction in the back‑arc.

The aftershocks were examined with the same four GMPEs. Their event terms are modestly positive (+0.15 to +0.20), but still fall within the worldwide scatter reported for Mw 6.5–7.0 interface events (±0.30). The intra‑event σ values for the aftershocks are larger (0.30–0.35) than the model‑specified sigma, indicating additional variability not captured by the equations. The authors attribute this excess to uncertainties in the VS30 proxies: many stations lack direct shear‑wave velocity measurements, and the study relied on regional averages or indirect estimates, inflating the site‑term uncertainty.

Spectral shape comparisons reveal that all models reproduce low‑frequency (≤0.2 Hz) motions well, where distance attenuation dominates. At higher frequencies (≥1 Hz) the divergence becomes evident, especially beyond 5 Hz where back‑arc recordings are 0.2–0.4 g lower than predicted. This behavior is consistent with the presence of shallow, low‑density sedimentary basins and more complex propagation paths in the back‑arc, which preferentially attenuate high‑frequency energy.

Overall, the paper reaches four principal conclusions: (1) Global and regional interface GMPEs are generally adequate for predicting ground‑motion amplitudes of the Ecuador megathrust and its aftershocks out to ~200 km; (2) Models that do not differentiate trench from back‑arc propagation (Zhao 2006, Montalva 2016) overestimate high‑frequency motions in the back‑arc and should be used with caution in that setting; (3) Accurate VS30 information is critical—uncertainties in site‑characterization substantially increase intra‑event variability and degrade model performance; (4) Future development of Ecuador‑specific (or broader South‑American) GMPEs should incorporate separate attenuation terms for trench and back‑arc paths, and should be calibrated with high‑quality VS30 measurements.

The study thus provides a valuable benchmark for seismic hazard assessments in Ecuador and neighboring subduction zones, demonstrating both the strengths of existing GMPEs and the specific regional adjustments required to improve ground‑motion forecasts for engineering and risk‑mitigation purposes.