The current default IM calculations are: PGA, PGV, CAV, AI, Ds575, Ds595, MMI, pSA, IESDR
The default periods pSA is calculated at are: 0.02, 0.05, 0.1, 0.2, 0.3, 0.4, 0.5, 0.75, 1.0, 2.0, 3.0, 4.0, 5.0, 7.5, 10.0
This can be extended into a logspace from 0.01 to 10 at 100 steps in base 10
An intermediate range of periods is to be determined for use in Seistech for UHS / empirical calculations
| Intensity Measure | Status | Notes |
|---|---|---|
| PGA/PGV/MMI | ||
| pSA | Cython - works well | |
| CAV | ||
| AI | ||
| Ds | Very Slow / unvectorised | |
| Advanced IMs | Current implemented models: SAC_Steel_MF_9Story, SAC_Steel_MF_3Story, Steel_MF_5Story To be implemented models: | |
| Taghavi and Miranda (TM05) | Proposed IM name is TM05_aX_cY_Ts_storeyL_calc X is alpha, Y is C, T is the period and I is the story specified. a = 0, 1.5, 5, 15, 30 c = 0.02, 0.05, 0.1 T = 0.1, 0.4, 0.7, 1.5, 3.0 calculations are disp_peak, slope_peak, storey_shear_peak, and total_accel_peak. In total there will be 3300 IMs calculated for a single waveform. This will be considered as an advanced IM. | |
| Fourier Spectra | Added, vectorised, but quite slow | |
| Directionality (via SA RotD100/RotD50) | https://slack-files.com/T0C0R97U5-FSBQH2RJB-5511f81cef – Boore NonGMmeasuresGMs_BSSA_2010 Added | |
| Ratio of elastic spectra to inelastic spectra | Same paper as above Matlab scripts received from Jack Baker, inelastic script ported to python and validated against the matlab script. Integrated into the workflow, IM names are IESDR_{period} where the same periods for pSA are used. | |
| period-to-period SA correlations | Future work (how do we validate with empirical) | |
Evolutionary IMs (Rezain et al 2015) (Evolution of intensity; Zero-level upcrossings, positive min & negative max) | (low priority; but noted here for completeness) |