Difference between revisions of "Accessing CyberShake Database Data"
(One intermediate revision by the same user not shown) | |||
Line 16: | Line 16: | ||
*Rupture_Variations - list of rupture variations (realizations with different hypocenters and slip distributions) for each rupture | *Rupture_Variations - list of rupture variations (realizations with different hypocenters and slip distributions) for each rupture | ||
*PeakAmplitudes - used to get the PSA values, given the run ID and a specific rupture, rupture variation, and period | *PeakAmplitudes - used to get the PSA values, given the run ID and a specific rupture, rupture variation, and period | ||
+ | *Hazard_Curves - contains the run id and IM type for each hazard curve | ||
+ | *Hazard_Curve_Points - the X and Y values describing each hazard curve | ||
The wiki page for each CyberShake study has the information - SGT ID, Rupture_Variation_Scenario_ID, etc. - which you need to retrieve the data from a specific study. | The wiki page for each CyberShake study has the information - SGT ID, Rupture_Variation_Scenario_ID, etc. - which you need to retrieve the data from a specific study. | ||
Line 27: | Line 29: | ||
and R.Max_Frequency=10.0 and R.ERF_ID=36 and R.Status="Verified" and R.SGT_Variation_ID=8 and R.Velocity_Model_ID=5 and R.Rup_Var_Scenario_ID=6 | and R.Max_Frequency=10.0 and R.ERF_ID=36 and R.Status="Verified" and R.SGT_Variation_ID=8 and R.Velocity_Model_ID=5 and R.Rup_Var_Scenario_ID=6 | ||
and P.Run_ID=R.Run_ID | and P.Run_ID=R.Run_ID | ||
− | and P.IM_Type_ID= | + | and P.IM_Type_ID=162 |
order by P.IM_Value desc limit 100; | order by P.IM_Value desc limit 100; | ||
The "R.Velocity_Model_ID=5" part means CVM-S4.26. The "R.Rup_Var_Scenario_ID=6" part means Graves & Pitarka (2014) rupture variations, with uniform hypocenter spacing. | The "R.Velocity_Model_ID=5" part means CVM-S4.26. The "R.Rup_Var_Scenario_ID=6" part means Graves & Pitarka (2014) rupture variations, with uniform hypocenter spacing. | ||
− | The "P.IM_Type_ID= | + | The "P.IM_Type_ID=162" means 3 second period for RotD50. This mapping is stored in the IM_Types table, but here are the values of most interest: |
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
− | |||
RotD50, Period = 10 sec, IM_Type_ID=152 | RotD50, Period = 10 sec, IM_Type_ID=152 | ||
Line 53: | Line 44: | ||
RotD50, Period = 0.2 sec, IM_Type_ID=174 | RotD50, Period = 0.2 sec, IM_Type_ID=174 | ||
RotD50, Period = 0.1 sec, IM_Type_ID=175 | RotD50, Period = 0.1 sec, IM_Type_ID=175 | ||
+ | |||
+ | RotD100 data has different IM Types; you could search the IM_Types table to identify the type IDs you want. Here are the values of most interest: | ||
RotD100, Period = 10 sec, IM_Type_ID=136 | RotD100, Period = 10 sec, IM_Type_ID=136 | ||
Line 63: | Line 56: | ||
RotD100, Period = 0.1 sec, IM_Type_ID=172 | RotD100, Period = 0.1 sec, IM_Type_ID=172 | ||
− | Duration data | + | We also have geometric mean data: |
+ | |||
+ | Period = 10 sec, IM_Type_ID=1 | ||
+ | Period = 5 sec, IM_Type_ID=11 | ||
+ | Period = 3 sec, IM_Type_ID=21 | ||
+ | Period = 2 sec, IM_Type_ID=26 | ||
+ | Period = 1 sec, IM_Type_ID=86 | ||
+ | Period = 0.5 sec, IM_Type_ID=88 | ||
+ | Period = 0.2 sec, IM_Type_ID=94 | ||
+ | Period = 0.1 sec, IM_Type_ID=99 | ||
+ | |||
+ | Duration data has different IM Types: | ||
acceleration significant duration, 5% to 75%, X component, IM_Type_ID=176 | acceleration significant duration, 5% to 75%, X component, IM_Type_ID=176 | ||
Line 76: | Line 80: | ||
In the above example, I have asked MySQL to order them in decreasing IM_Value, so the largest PSA values are at the top. Since the query would return about 400,000 values, I've also added the 'limit 100' so I only get the top 100 values. | In the above example, I have asked MySQL to order them in decreasing IM_Value, so the largest PSA values are at the top. Since the query would return about 400,000 values, I've also added the 'limit 100' so I only get the top 100 values. | ||
− | If you run this query, you should get source 244, rupture 12, rupture variation | + | If you run this query, you should get source 244, rupture 12, rupture variation 25 as your largest RotD50 PSA at 3 seconds. If you wanted to learn more about this rupture, you could run the following query: |
select * from Ruptures | select * from Ruptures |
Latest revision as of 19:54, 28 October 2021
To query the CyberShake database, you can use the following login information with a MySQL client:
host: focal.usc.edu user: cybershk_ro password: CyberShake2007 database: CyberShake
This username/password is read-only, so you don't need to worry about accidentally changing the database.
We have quite a few tables and quite a lot of information in the database, but there are only a few tables that you'll need to use. They are:
- CyberShake_Sites - used to look up the unique site ID from the name of the site (like "LADT").
- CyberShake_Runs - used to look up the unique run ID using a site ID and other run parameters (like the velocity model, broadband, etc.)
- Ruptures - list of ruptures from UCERF used in CyberShake, along with their magnitude, name, and probability.
- CyberShake_Site_Ruptures - list of ruptures used for each site, including the distance to the rupture
- Rupture_Variations - list of rupture variations (realizations with different hypocenters and slip distributions) for each rupture
- PeakAmplitudes - used to get the PSA values, given the run ID and a specific rupture, rupture variation, and period
- Hazard_Curves - contains the run id and IM type for each hazard curve
- Hazard_Curve_Points - the X and Y values describing each hazard curve
The wiki page for each CyberShake study has the information - SGT ID, Rupture_Variation_Scenario_ID, etc. - which you need to retrieve the data from a specific study.
So if you'd like to get all the PSA data for LADT for 3 seconds from Study 15.12, you'd want to run the following query:
select P.Source_ID, P.Rupture_ID, P.Rup_Var_ID, P.IM_Value from CyberShake_Sites S, CyberShake_Runs R, PeakAmplitudes P where S.CS_Short_Name="LADT" and S.CS_Site_ID=R.Site_ID and R.Max_Frequency=10.0 and R.ERF_ID=36 and R.Status="Verified" and R.SGT_Variation_ID=8 and R.Velocity_Model_ID=5 and R.Rup_Var_Scenario_ID=6 and P.Run_ID=R.Run_ID and P.IM_Type_ID=162 order by P.IM_Value desc limit 100;
The "R.Velocity_Model_ID=5" part means CVM-S4.26. The "R.Rup_Var_Scenario_ID=6" part means Graves & Pitarka (2014) rupture variations, with uniform hypocenter spacing.
The "P.IM_Type_ID=162" means 3 second period for RotD50. This mapping is stored in the IM_Types table, but here are the values of most interest:
RotD50, Period = 10 sec, IM_Type_ID=152 RotD50, Period = 5 sec, IM_Type_ID=158 RotD50, Period = 3 sec, IM_Type_ID=162 RotD50, Period = 2 sec, IM_Type_ID=167 RotD50, Period = 1 sec, IM_Type_ID=169 RotD50, Period = 0.5 sec, IM_Type_ID=173 RotD50, Period = 0.2 sec, IM_Type_ID=174 RotD50, Period = 0.1 sec, IM_Type_ID=175
RotD100 data has different IM Types; you could search the IM_Types table to identify the type IDs you want. Here are the values of most interest:
RotD100, Period = 10 sec, IM_Type_ID=136 RotD100, Period = 5 sec, IM_Type_ID=142 RotD100, Period = 3 sec, IM_Type_ID=146 RotD100, Period = 2 sec, IM_Type_ID=151 RotD100, Period = 1 sec, IM_Type_ID=168 RotD100, Period = 0.5 sec, IM_Type_ID=170 RotD100, Period = 0.2 sec, IM_Type_ID=171 RotD100, Period = 0.1 sec, IM_Type_ID=172
We also have geometric mean data:
Period = 10 sec, IM_Type_ID=1 Period = 5 sec, IM_Type_ID=11 Period = 3 sec, IM_Type_ID=21 Period = 2 sec, IM_Type_ID=26 Period = 1 sec, IM_Type_ID=86 Period = 0.5 sec, IM_Type_ID=88 Period = 0.2 sec, IM_Type_ID=94 Period = 0.1 sec, IM_Type_ID=99
Duration data has different IM Types:
acceleration significant duration, 5% to 75%, X component, IM_Type_ID=176 acceleration significant duration, 5% to 95%, X component, IM_Type_ID=177 velocity significant duration, 5% to 75%, X component, IM_Type_ID=178 velocity significant duration, 5% to 95%, X component, IM_Type_ID=179 acceleration significant duration, 5% to 75%, X component, IM_Type_ID=180 acceleration significant duration, 5% to 95%, X component, IM_Type_ID=181 velocity significant duration, 5% to 75%, X component, IM_Type_ID=182 velocity significant duration, 5% to 95%, X component, IM_Type_ID=183
In the above example, I have asked MySQL to order them in decreasing IM_Value, so the largest PSA values are at the top. Since the query would return about 400,000 values, I've also added the 'limit 100' so I only get the top 100 values.
If you run this query, you should get source 244, rupture 12, rupture variation 25 as your largest RotD50 PSA at 3 seconds. If you wanted to learn more about this rupture, you could run the following query:
select * from Ruptures where ERF_ID=36 and Source_ID=244 and Rupture_ID=12;
You'll see it's a Puente Hills rupture, with magnitude 7.25 and probability ~1.97e-06. The probability of the rupture is divided evenly among its rupture variations.
If you wanted to get the distance from LADT to this rupture, you could run the following query:
select SR.Site_Rupture_Dist from CyberShake_Site_Ruptures SR, CyberShake_Sites S where S.CS_Short_Name="LADT" and S.CS_Site_ID=SR.CS_Site_ID and SR.ERF_ID=36 and SR.Source_ID=244 and SR.Rupture_ID=12;
This will give you the shortest distance from the site to any part of the rupture surface, in km.