- Ground-motion model is:
where PGA is in cm∕s2, a1 = 0.5586, a2 = -1.0902, a3 = -0.0035 and σA = 0.2030 (intra-event) and
σE = 0.2006 (inter-event). Derive individual Cst values for each of the 35 stations (not reported here).
- Focal depths 0.1 ≤ H ≤ 226.3km. 10 events in Bucaramanga nest.
- Also derive models using functional forms of García et al. (2005) and Joyner and Boore (1981) (using
- Use data from networks in Medellėn (1997–2010) and Aburrā Valley (2008–2010). Medellėn array has had
27 stations of which 22 (3 on rock, 1 at 40m depth) are still operating. 13 stations with Etna Episensors, 1
with K2, 1 FBA-23DH (in borehole) and 8 CUSP-3C. Aburrā Valley array has 11 surface stations (9 with
CUSP-3C and 2 Etna Episensors). Do not use data from 1 station because it did not record any selected
- Select events that were recorded at ≥ 40% (15 of 36) of stations.
- Baseline correct records. Examine signal-to-noise ratios (3s around peak to 3s before P-wave) and conclude
high enough to estimate PGA.
- Strong positive correlation between ML and repi.
- Compute residuals w.r.t. model, plot residuals w.r.t repi and also use approach of Scherbaum et al. (2004)
to examine fit of candidate models to observations. Find that model reported above performs best.