Ground Acceleration PGA Calculator

Peak ground acceleration is the largest absolute value in the acceleration time history recorded at a site during an earthquake, normally quoted as a fraction of g or in Gal, where 1 Gal is 1 cm/s2 and 1 g equals 980.7 Gal. It is estimated with a ground motion prediction equation, a regression fitted to thousands of recordings, linear in magnitude and logarithmic in distance: ln(PGA) = c1 + c2(M-6) + c3(M-6)^2 + c4 ln(Reff) + c5 R. The c4 term captures geometric spreading, c5 captures anelastic attenuation from inelastic absorption along the path, and Reff = sqrt(R^2 + h^2) keeps the prediction finite as distance approaches zero by embedding a fictitious depth of a few kilometres. Local soil is handled with a site factor keyed to NEHRP classes, which are defined by average shear-wave velocity in the top 30 m: class A above 1,500 m/s, B from 760 to 1,500, C from 360 to 760, D from 180 to 360 and E below 180, with amplification rising from roughly 0.8 to 2.5 as the profile softens. PGA maps onto felt intensity roughly, with 0.04 g at the damage threshold, 0.09 to 0.34 g for strong shaking and above 0.34 g for severe; codes convert it into design spectral values, taking SDS as two-thirds of the site-modified short-period spectral acceleration. Two limitations dominate: PGA is a single high-frequency peak that correlates poorly with damage to long-period structures, and GMPE scatter is large, with a standard deviation in natural-log terms near 0.5, a factor of roughly 1.6 either way at one sigma.

Instrumental strong-motion recording began with the accelerographs the US Coast and Geodetic Survey deployed in southern California from 1932. Their first significant records came from the Long Beach earthquake of 10 March 1933, and the 1940 Imperial Valley event produced the El Centro record, with a peak near 0.35 g, which served as the reference input for structural design and research for decades afterward. Before instruments existed, shaking was ranked by observed effects using the scale Giuseppe Mercalli published in 1902, which Harry Wood and Frank Neumann adapted to American construction in 1931 as the Modified Mercalli Intensity scale. The 1971 San Fernando earthquake yielded a Pacoima Dam record above 1.2 g and, together with hundreds of other recordings, made statistical attenuation modelling practical. Allin Cornell set out probabilistic seismic hazard analysis in 1968, giving PGA a formal role in computing exceedance probabilities, and William Joyner and David Boore published influential empirical attenuation relations in 1981. The Pacific Earthquake Engineering Research Center coordinated the Next Generation Attenuation project, releasing the NGA-West models in 2008 and NGA-West2 in 2014, which added terms for hanging-wall geometry, basin depth and rupture directivity. David Wald and colleagues published PGA-to-intensity conversions in 1999 that underpin USGS ShakeMap, and mapped PGA at a 2 percent chance of exceedance in 50 years became the basis of United States seismic design provisions.