Seismosoft | Earthquake Engineering Solutions

You have correctly identified one of the outstanding issues with maximum direction (RotD100-based) scaling of ground motions. With GeoMean-based scaling, you amplify both components be the computed scale factor. With RotD100-based scaling, you still amplify both components by the computed scale factor. You still do not know the correct angle. It has been shown (Importance of Target Spectrum Basis in Earthquake Ground Motion Scaling, Huff, 2020) that a RotD100-based scale factor is approximately the same as a GeoMean-based scale factor. That is to say, if you match the RotD100 of the ground motion to the RotD100-based design response spectrum, you should get about the same scale factor as if you had matched the geometric mean of the ground motion record to the GeoMean-based design response spectrum. So, as currently defined, RotD100-based scaling seems to do little more than make the computation more difficult. I can see the benefit of RotD100-based scaling for 2D analysis - you just match one component to the RotD100 spectrum. For 3D analysis, GeoMean-based scaling makes more sense, at least as I understand it. I could be mistaken. Hope this helps.

Thank you so much dear huffte,
I read your helpful article (Importance of Target Spectrum Basis in Earthquake Ground Motion Scaling, Huff, 2020), and because I want to use ASCE7-22 MCE spectrum which I have from USGS as "target spectrum" so I should use RotD100 for scaling time histories?
and then just multiply each H1 and H2 component by scale factor?
thanks a lot

Be careful if you get the data from USGS. While virtually all data in codes (AASHTO and ASCE 7) are based on USGS data, there are different bases available.

If you use ASCE 7-22 based USGS data from here, you will get RotD100, Risk-targeted ground motion data, and you should scale records RotD100 to the design response spectrum obtained:

https://earthquake.usgs.gov/ws/designmaps/

Dear Mr. Huff
Yes as you said I use ASCE7-22 based USGS data but the site which I use to get Multi-Period MCE Spectrum is below.
So the Multi-Period MCE Spectrum which I get from this site is RotD100, and I should use RotD100 for scaling records. Am I correct?

https://asce7hazardtool.online/

thank you so much

Yes. ASCE 7-22 data are RotD100. You compute RotD100 for the ground motion record pair and scale to the design response spectrum. I quote from ASCE 7-22, Section 16.2.3.2:

"For each horizontal ground motion pair, a maximum-direction spectrum shall be constructed from the two horizontal ground motion components. Each ground motion shall be scaled, with an identical scale factor applied to both horizontal components, such that the average of the maximum-direction spectra from all ground motions generally matches or exceeds the target response spectrum over the period range defined in Section 16.2.3.1. The average of the maximum-direction spectra from all the ground motions shall not fall below 90% of the target response spectrum for any period within the same period range."

Dear Mr. Huff
Thank you so much for your help and time.