How does spectral matching improve the accuracy of NRHA when selecting ground motion records?

Spectral matching improves the accuracy of Nonlinear Response History Analysis (NRHA) when selecting ground motion records by modifying the selected records to closely match a target response spectrum over a specified period range. This process ensures that the ground motions used in the analysis are more representative of the seismic hazard at the site for the specific structure being analyzed. A target response spectrum, often derived from a Uniform Hazard Spectrum (UHS) or Conditional Mean Spectrum (CMS), represents the expected ground motion intensity at different periods for a given probability of exceedance. Spectral matching algorithms adjust the frequency content and amplitude of the original ground motion record to minimize the difference between its response spectrum and the target spectrum. This is typically achieved by adding wavelets or modifying the time-frequency characteristics of the record. By ensuring that the ground motion records closely match the target spectrum, spectral matching reduces the variability in the structural response caused by using records with different frequency content. This leads to more consistent and reliable estimates of the structural demands, such as displacements, forces, and ductility demands. It's important to note that excessive spectral matching can artificially alter the characteristics of the ground motion and may reduce its realism. Therefore, spectral matching should be performed carefully, and the original characteristics of the ground motion records should be considered.