The Ground Motion Prediction Equation (GMPE) is very important in estimating the intensity of earthquake shocks as a basis for risk mitigation. This study aims to modify and validate the Akkar & Boomer (2007) GMPE using shallow earthquake data in the North Sumatra region for the period 2017–2023. The earthquake data were obtained from BMKG and included parameters such as magnitude, depth, and distance from the source. The analysis method involved nonlinear regression, data cleaning, and validation using residual analysis. The results showed that the maximum ground acceleration (PGA) tended to decrease nonlinearly with increasing distance from the earthquake source. The modified GMPE equation was: Log₁₀ PGA = −0.5916 + 0.5875M + 0.0576M² + (−0.8699 + − 0.1985M) Log₁₀(√R² + 8.2032²) + 0.105, with an R² value of 0.56 and prediction error values such as 0.21; MAE 0.36; RMSE 0.46; STD 0.46). Thus, the modification of GMPE based on local data can provide a more representative estimate of earthquake hazards to support mitigation efforts in North Sumatra.
Copyrights © 2025