In the current literature, several procedures for computing average horizontal to vertical (H/V) spectral ratio of ambient vibrations are interchangeably adopted, whose possible equivalence has never been checked. To this purpose, expected sampling properties of average H/V spectral ratio relative to commonly adopted procedures are determined in the frame of a coherent physical–statistical model. This analysis reveals that all the considered estimators provide H/V estimates that are biased of amounts ranging from 46 per cent up to above 100 per cent of the true value. The amount of these biases is not reduced when H/V spectral ratio estimates are determined, as in the common practice, by averaging H/V spectral ratios computed over a number of time-windows extracted from the experimental record. On the other hand, these biases can be drastically reduced in the case that H/V spectral ratio is computed as the squared root of the ratio of horizontal (whatever the north/south–east/west
Albarello, D., Lunedei, E. (2013). Combining horizontal ambient vibration components for H/V spectral ratio estimates. GEOPHYSICAL JOURNAL INTERNATIONAL, 194(2), 936-951 [10.1093/gji/ggt130].
Combining horizontal ambient vibration components for H/V spectral ratio estimates
Albarello, D.;Lunedei, E.
2013-01-01
Abstract
In the current literature, several procedures for computing average horizontal to vertical (H/V) spectral ratio of ambient vibrations are interchangeably adopted, whose possible equivalence has never been checked. To this purpose, expected sampling properties of average H/V spectral ratio relative to commonly adopted procedures are determined in the frame of a coherent physical–statistical model. This analysis reveals that all the considered estimators provide H/V estimates that are biased of amounts ranging from 46 per cent up to above 100 per cent of the true value. The amount of these biases is not reduced when H/V spectral ratio estimates are determined, as in the common practice, by averaging H/V spectral ratios computed over a number of time-windows extracted from the experimental record. On the other hand, these biases can be drastically reduced in the case that H/V spectral ratio is computed as the squared root of the ratio of horizontal (whatever the north/south–east/westFile | Dimensione | Formato | |
---|---|---|---|
lunedei.pdf
non disponibili
Tipologia:
PDF editoriale
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
674.7 kB
Formato
Adobe PDF
|
674.7 kB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/11365/45946
Attenzione
Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo