Gupta 2010

class openquake.hazardlib.gsim.gupta_2010.Gupta2010SSlab[source]

Implements GMPE of Gupta (2010) for Indo-Burmese intraslab subduction.

This model is closely related to the model of Atkinson & Boore (2003). In particular the functional form and coefficients C2-C7 of Gupta (2010) are adopted from Atkinson & Boore (2003). The only substantive changes are a) the horizontal component modeled is different (as noted below) b) a coefficient C8 and a dummy variable v are added to model vertical motion and c) the coefficient C1 is recalculated based on a database of “a total of 56 three-component accelerograms at 37 different sites from three in-slab earthquakes along the Indo-Burmese subduction zone” (p 370).

Equation (2) p. 373 gives the form of the equation which was fitted:

log Y - C2*M - C3*h - C4*R + g log R = C1 + C8*v + sigma

The left-hand side of this equation was computed using event parameters and the coefficients of Atkinson & Boore (2003). The regression coefficients C1 and C8 on the right-hand side were slightly smoothed after fitting. Note that since “v=0 for horizontal and 1 for vertical motion”, and since the current implementation only models horizontal motion, we can subclass directly from openquake.hazardlib.gsim.atkinson_boore_2003.AtkinsonBoore2003SSlab, modifying only the metadata constants and regression coefficients.

Page number citations in this documentation refer to Gupta (2010).


Gupta, I. (2010). Response spectral attenuation relations for in-slab earthquakes in Indo-Burmese subduction zone. Soil Dyn. Earthq. Eng., 30(5):368–377.

Atkinson, G. M. and Boore, D. M. (2003). Empirical ground-motion relations for subduction-zone earthquakes and their application to Cascadia and other regions. Bull. Seism. Soc. Am., 93(4):1703–1729.

COEFFS_SSLAB = <openquake.hazardlib.gsim.base.CoeffsTable object>

Coefficients taken from Table 3, p. 884. The row for 0.02 was relabeled PGA since the paper indicates this is what it actually is (see p. 371) and since these were the coefficients for PGA in Atkinson & Boore (2003).


Unlike Atkinson & Boore (2003), “rather than the random horizontal component, the geometric mean of both the horizontal components has been used in the modified attenuation relations.” (p. 376)

DEFINED_FOR_INTENSITY_MEASURE_TYPES = set([<class 'openquake.hazardlib.imt.PGA'>, <class 'openquake.hazardlib.imt.SA'>])

“The actual peak ground acceleration (PGA) from the corrected time histories are taken as the response spectral amplitudes at a period of 0.02 s (50 Hz frequency).” p. 371. Based on this comment, the coefficients labeled as being for 0.02 s have been relabeld as PGA.


Since the database is small only the total standard deviation is reported.


As stated in the title.

non_verified = True

Mean value data obtained from author matched well at 1 s and below but not at longer periods. As a temporary measure the reference test result has been generated from the current implementation.