Maximum Energy and Total Fault Slip Estimate of the NVT during the 2006 SSE To estimate an upper edge of seismic energy released by all nonvolcanic tremor episodes (NVT) and the entire hypothetical fault slip produced by the NVT during the 2006 slow slip event (SSE) we assume isotropic S wave energy in a uniform medium. Despite this simplification we only need to find an extreme upper limit of the NVT radiated seismic energy, Es, during the period of 2006 SSE. The maximum of NVT total coseismic slip may be estimated from Es assuming that all NVT episodes were occurring on the same interplate thrust fault with the same rake. We used the algorithm for radiated seismic energy [Battaglia and Aki, 2003; Maeda and Obara, 2009]. The basic equation is Es=2pi·p·ß·R^2·exp(2pi·fc·R/ß·Q)·Integral(t1->t2)[Un(t)^2+Ue(t)^2+Uz(t)^2]dt , (1) where Es is the radiated seismic energy, p is the density (2800 kg/cm3), ß is the shear velocity (3500 m/s), R is the distance between the NVT source and a seismic station, U is the velocity amplitude of the each component, fc is the center frequency and Q is the quality factor. The amplitude, U, was summed over an NVT episode lasting from t1 to t2 according to the NVT duration register. When a background seismic spectrum (seismic records with no NVT or earthquakes) was removed from the seismic spectrum containing the NVT, the result was the NVT had a peak at ~1.5 Hz (Fig. S1), so this was used as fc. Q = 276 as determined [Garcia et al., 2004] in central Mexico. We chose R = 65 km (> sqrt(40^2+50^2)) as a reasonable upper limit on the distance of detected NVT, where 40 km was assumed to be the maximum possible depth and 50 km was the maximum horizontal distance of reliable tremor detection. A depth of about 40 km would put all NVT at the interplate interface. Nearly all of the MASE tremor records show the vertical component much weaker than the horizontal components, which means that the detected NVT was located close (less than ~20 km) or almost directly below the seismic stations, suggesting a relatively short horizontal distance from the MASE profile. However, some NVT were located more distantly from the MASE profile [Payero et al., 2008]. The NVT had been located within a band of ~100 km along the MASE profile [Payero et al., 2008]. The present study shows that the same distance range contains almost all of the NVT energy (Fig. 2). The edge of the detectable tremor distance was ~50 km, from the NVT peak energy where the NVT energy was more than an order of magnitude lower than at the peak (Fig. S2). Thus we assumed that the maximum horizontal distance of the NVT was 50 km from the MASE line, and the entire fault area of the recorded tremor was about of 10^4 km2. The seismic records of NVT were filtered in the band of 1-2 Hz where the signal to noise ratio was the best for all stations ([Payero et al., 2008], and Fig. S1). In order to calculate Es, it is necessary to know the ratio e of the energy from the total NVT (noise subtracted) spectrum (Figure S1) to the 1-2 Hz energy band, dEs. This ratio estimated from the spectrum in Figure S1 is approximately e~10. The upper limit for radiated seismic energy, dEs, estimated using (1), in the frequency range of 1-2Hz for all detected NVT during the 2006 SSE was dEs=4.72x10^8 J. Multiplying e to dEs gives Es=4.72x10^9 J (Me~3.3, [Gutenberg and Richter, 1956]) which is well below the total energy of the SSE (Mw ~7.0). The total fault area where all recorded NVT energy, Es, was radiated is about of A=100x100 km2. NVT stress drop, dS, may vary if the tremor source has a characteristic scale [Watanabe et al., 2007] but it is apparently as low as the amplitude of the NVT triggering stress dS < 0.4-0.6 MPa [Peng and Chao, 2008; Rubinstein et al., 2007]. So a rough estimate of the maximum cumulative NVT fault slip is about d = 2Es/(A dS)~2x10^-3 mm. This dislocation on the 100 km long plate interface at the depth of 40 km would produce a negligible surface deformation [Savage, 1983], which is at least two orders of magnitude less than the noise level of the GPS position time series. We should remind that these upper edge estimates of ES and d would be appropriate only if all the above taken assumptions were valid. In reality the values of ES and d may be significantly lower. References Battaglia, J., and K. Aki (2003), Location of seismic events and eruptive fissures on the Piton de la Fournaise volcano using seismic amplitudes, J. Geophys. Res., 108. Garcia, D., S. K. Singh, M. Herraiz, J. F. Pacheco, and M. Ordaz (2004), Inslab Earthquakes of Central Mexico: Q, Source Spectra, and Stress Drop, B Seismol Soc Am, 94(3), 789-802. Maeda, T., and K. Obara (2009), Spatiotemporal distribution of seismic energy radiation from low-frequency tremor in western Shikoku, Japan, J. Geophys. Res., 114. Payero, J. S., V. Kostoglodov, N. Shapiro, T. Mikumo, A. Iglesias, X. Perez-Campos, and R. W. Clayton (2008), Nonvolcanic tremor observed in the Mexican subduction zone, Geophys. Res. Lett., 35(7). Peng, Z., and K. Chao (2008), Non-volcanic tremor beneath the Central Range in Taiwan triggered by the 2001 Mw 7.8 Kunlun earthquake, Geophys J Int, 175(2), 825-829. Rubinstein, J. L., J. E. Vidale, J. Gomberg, P. Bodin, K. C. Creager, and S. D. Malone (2007), Non-volcanic tremor driven by large transient shear stresses, Nature, 448(7153), 579-582. Savage, J. C. (1983), A Dislocation Model of Strain Accumulation and Release at a Subduction Zone, J. Geophys. Res., 88(B6), 4984-4996. Watanabe, T., Y. Hiramatsu, and K. Obara (2007), Scaling relationship between the duration and the amplitude of non-volcanic deep low-frequency tremors, Geophys. Res. Lett., 34.