Modified Mercalli Intensity (MMI) Scale and ShakeMaps
Every earthquake has one magnitude but produces many different levels of shaking. This shaking is typically described using the Modified Mercalli Intensity Scale, or MMI. The MMI scale is a qualitative measure of the effects and intensity of an earthquake at a specific location based on observed impacts on people, as well as the built and natural environment. The scale ranges from I to XII, with level I indicating that the earthquake was imperceptible by most people and level XII signifying destruction. Historically, shaking intensity was determined from subjective reports of how people experienced the earthquake and observed damage in specific locations. These intensity reports from areas near the earthquake epicenter were used to create maps showing the distribution and severity of shaking and damage.
Today, intensity maps, commonly known as ShakeMaps (Worden et al., 2020), are automatically generated within minutes of an earthquake using data collected by seismic instruments. ShakeMaps provide a detailed geographic overview of an earthquake's effects and serve as a critical resource for emergency response, damage assessment, and community impact. The CGS Cumulative Shaking Intensity map (opens new tab) compiles all ShakeMaps generated from 1981 to 2023, offering valuable insights into historical earthquake impacts.
MMI in Map Sheet 48
Instead of reflecting shaking from past earthquakes, the CGS MS 48 illustrates anticipated shaking intensity from potential future earthquakes. This anticipated shaking intensity is converted from ground motions calculated using the 2023 update of the United States Geological Survey (USGS) National Seismic Hazard Model (NSHM) (Petersen et. al., 2023, Powers et al., 2022). MS 48 is typically revised following each major update of the NSHM. Earlier versions of this map depicted relatively long-period (1.0 second) SA, as it correlates well with overall earthquake damage. However, with the 2023 update of the NSHM, shaking intensity is now represented using the MMI scale, as it is more intuitive and easier for most people to understand than the 1.0-second SA.
The MMI values depicted in MS 48 are converted from PGV using the GMICE of Worden et al. (2012). Worden et al. also established relations between MMI and other ground motion parameters, including PGA, and 0.3-s, 1.0-s, and 3.0-s SAs. Among these, the relationship with PGV yields the smallest errors. Although Worden et al. also explored linear combinations of PGA and PGV, as well as equations incorporating magnitude and distance, these provided only nominally improved results (less than 0.1 intensity unit). For simplicity, MS 48 uses MMI values derived solely from PGV. Additionally, PGV was selected because its distribution across California closely resembles the distribution of 1.0-second SA from previous versions of the map. Both parameters are more sensitive to characteristics of shallow geologic materials than PGA or short-period SAs, effectively capturing the amplified shaking experienced in sedimentary basins under otherwise similar conditions.
Other MMI Maps
A map of MMI from PGV at 10% in 50 years hazard level, along with maps of MMI converted using PGA, are presented in the “MMI Hazard Maps” section for comparison. The GMICEs used for MMI calculation in these maps are summarized in Table 1. It is important to note the significant uncertainty involved in MMI conversion, as demonstrated by notable differences in MMIs calculated using various GMICEs from Worden et al. (2012). To further highlight this uncertainty, additional maps were produced using more recent GMICEs developed by Gallahue and Abrahamson (2023) (see the last two equations in Table 1). These newer equations are based on PGA and incorporate either the number of standard deviations (ɛ) of the PGA or magnitude (M) and distance (R). The parameters M, R, and ɛ are derived from hazard disaggregation. As shown in Table 1, the resulting MMIs are labeled as MMI_pgv, MMI_pga, MMI_ɛ, and MMI_MR, respectively.
Table 1. Summary of the ground motion to intensity conversion equations used by the CGS
| MMI_pgv |
MMI = 3.78 + 1.47log(PGV) for log(PGV) ≤ 0.53
MMI = 2.89 + 3.16log(PGV) for log(PGV) > 0.53 |
Worden et al. (2012) |
| MMI_pga |
MMI = 1.78 + 1.55log(PGA) for log(PGA) ≤ 1.57
MMI = -1.60 + 3.70log(PGA) for log(PGA) > 1.57 |
| MMI_pga_ ɛ |
I = 8.622+1.230ln(PGA)+0.056(ln(PGA)-ln(0.1))2-0.568ɛln(PGA) |
Gallahue and Abrahamson (2023) |
| MMI_pga_MR |
I = 2.919 + 0.356ln(PGA)+0.01(ln(PGA)-ln(0.1))2 + 1.041M -0.889ln(R) |
|
Table 1 footnotes
- MMI_pgv: MMI calculated from peak ground velocity (PGV); MMI_pga: MMI calculated from peak ground acceleration (PGA), MMI_pga_ɛ: MMI calculated from PGA and disaggregated ground motion variability level in ln(PGA), MMI_pga_MR: MMI calculated from PGA and disaggregated mean magnitude (M) and distance (R)
- See “References” section for details.
|
Ground Motion Hazard Calculation
We used the USGS hazard code, nshmp-haz (opens new tab) (version 2.0.0; Powers et al. 2022) to perform ground motion hazard calculations. The seismic source model applied in our analysis is the 2023 USGS CONUS (contiguous United States) model, nshm-conus-6.0.0 (opens new tab). By default, the USGS CONUS model accounts for basin effects in the Los Angeles basin, the San Francisco basin, and the Great (Central) Valley (Petersen et al., 2023).
A key difference between our hazard maps and those published by USGS (Petersen et al., 2023) is that the USGS hazard maps assume uniform site condition, while our calculation accounts for variable site conditions. In hazard calculation, site condition is approximately represented by Vs30 (the average shear wave velocity in the top 30 meters of the Earth's crust). For our maps, Vs30 values at grid points were extracted from the July 2022 version of the California Vs30 map developed by Thompson (2018) (opens new tab). This map, reproduced in the “Additional Maps” section, is the best Vs30 map currently available for California. It incorporates the geology-based Vs30 data by Wills et al. (2015), which was also used in the 2016 version of MS 48.
The “Additional Hazard Maps” section includes a Map of Historical Earthquakes in California since 1769 and a Map of California Faults and Slip Rates for the 2023 NSHM, which illustrate some of the data used in ground motion hazard calculations. Earthquake data for the Historical Earthquakes map are sourced from the USGS ANSS Comprehensive Earthquake Catalog (opens new tab), with events scaled by magnitude and color-coded. The Faults and Slip Rates map uses data from Hatem et al. (2022 (opens new tab), Figure 7a), which presents the preferred slip rate values derived from the geologic deformation model. In this map, faults are color-coded according to their preferred slip rate values. In addition to seismicity and fault data, the NSHM also incorporates crustal deformation data and integrates the latest scientific models that predict fault ruptures and ground motions (Petersen et al., 2023).
Hazard curves for PGA, PGV, SAs were calculated for individual grid points across a uniform grid covering the State of California, with a grid spacing of 0.01 degrees (approximately 1 km). Map data for these ground motion parameters, representing the 10% and 2% chances of being exceeded in 50 years, were interpolated from hazard curves. Maps for some of these ground motion parameters are presented in the “Ground Motion Hazard Maps” section.
MMI Calculation
Hazard curves for PGV and PGA were used to calculate the MMI_pgv and MMI_pga hazard curves, respectively, using the GMICEs of Worden et al. (2012) (the first two equations in Table 1). Map data for MMI_pgv and MMI_pga, corresponding to the 10% and 2% chances of exceedance in 50 years, were then interpolated from these hazard curves. The calculation of MMI_ɛ and MMI_MR, however, required disaggregation of probabilistic hazards at the specified hazard levels (i.e., 2% and 10% in 50 years) to obtain the mean ɛ, M, and R, which demanded significantly more computational resources. Disaggregation was performed on an upgraded CGS computer using the disaggregation module of the USGS nhsmp-haz (opens new tab) code at the same 0.01-degree grid spacing used in ground motion hazard calculations. The resulting disaggregation data, in JSON file format, were parsed and used for MMI calculations.
Data
Data for all MMI and ground motion maps presented on this web page can be downloaded in the “Datafiles Download” section below.
