Study 15.12 Data Products: Difference between revisions

From SCECpedia
Jump to navigationJump to search
← Older editNewer edit →
Scottcal (talk | contribs)
8,097 edits
Scottcal (talk | contribs)
8,097 edits
Line 95: Line 95:


=== Changing RVSIG ===
=== Changing RVSIG ===
We reran PAS using rvsig1 = 0.1; comparison curves are below.
We reran PAS using rvsig1 = 0.1; comparison curves are below.


Line 102: Line 101:
| [[File:PAS_Run4667_v_4282_0.2sec_GEOM.png|400px|thumb|left|Blue: rvsig1=0.1, black: rvisg1=-1.0]]
| [[File:PAS_Run4667_v_4282_0.2sec_GEOM.png|400px|thumb|left|Blue: rvsig1=0.1, black: rvisg1=-1.0]]
|}
|}
=== Disaggregations ===
We did disaggregations for CyberShake and the 4 NGA-2s at 0.5 sec and 0.2 sec, at prob=1e-4.  Below are images for each, along with the top 5 contributing sources:
{|
! Data Source
! 0.5 sec value at prob=4e-4
! 0.5 sec plot
! 0.5 sec contributors
! 0.2 sec value at prob=4e-4
! 0.2 sec plot
! 0.2 sec contributors
|-
! CyberShake
| 1.08g
| [[File:PAS_Run4282_Disagg_4e-4_0.5sec_CyberShake.png|400px|thumb|left]]
| 47.0% Verdugo (source 283)
  23.4% Elysian Park (Upper) (source 158)
  12.5% Raymond (source 247)
  10.6% Puente Hills (LA) (source 244)
  5.9% Puente Hills (source 242)
| [[File:PAS_Run4282_Disagg_4e-4_0.2sec_CyberShake.png|400px|thumb|left]]
|
|-
! ASK2014
| 1.24g
| [[File:PAS_Run4282_Disagg_4e-4_0.5sec_ASK2014.png|400px|thumb|left]]
|
| [[File:PAS_Run4282_Disagg_4e-4_0.2sec_ASK2014.png|400px|thumb|left]]
|
|-
! BSSA2014
| 1.51g
| [[File:PAS_Run4282_Disagg_4e-4_0.5sec_BSSA2014.png|400px|thumb|left]]
|
| [[File:PAS_Run4282_Disagg_4e-4_0.2sec_BSSA2014.png|400px|thumb|left]]
|
|-
! CB2014
| 1.59g
| [[File:PAS_Run4282_Disagg_4e-4_0.5sec_CB2014.png|400px|thumb|left]]
|
| [[File:PAS_Run4282_Disagg_4e-4_0.2sec_CB2014.png|400px|thumb|left]]
|
|-
! CY2014
| 1.65g
| [[File:PAS_Run4282_Disagg_4e-4_0.5sec_CY2014.png|400px|thumb|left]]
|
| [[File:PAS_Run4282_Disagg_4e-4_0.2sec_CY2014.png|400px|thumb|left]]
|
|-

Revision as of 18:17, 25 May 2016

Hazard Maps

Below are hazard maps from Study 15.12 (150 ppi).

Basic Maps

0.2 sec geometric mean
0.2 sec geometric mean hazard map, all sites
0.2 sec geometric mean hazard map, all sites, marked
0.2 sec 2008 NGA basemap
0.5 sec geometric mean
0.5 sec geometric mean hazard map, all sites
0.5 sec geometric mean hazard map, all sites, marked
0.2 sec 2008 NGA basemap
1 sec geometric mean
1 sec geometric mean hazard map, all sites
1 sec geometric mean hazard map, all sites, marked
0.2 sec 2008 NGA basemap
2 sec geometric mean
2 sec geometric mean hazard map, all sites
2 sec geometric mean hazard map, all sites, marked
3 sec geometric mean
3 sec geometric mean hazard map, all sites
3 sec geometric mean hazard map, all sites, marked
5 sec geometric mean
5 sec geometric mean hazard map, all sites
5 sec geometric mean hazard map, all sites, marked
10 sec geometric mean
10 sec geometric mean hazard map, all sites
10 sec geometric mean hazard map, all sites, marked

Comparison Maps

2 sec 3 sec 5 sec 10 sec
Difference
Study 15.12 vs Study 15.4, 2 sec geometric mean, difference map. Warm colors are higher Study 15.12.
Study 15.12 vs Study 15.4, 3 sec geometric mean, difference map. Warm colors are higher Study 15.12.
Study 15.12 vs Study 15.4, 5 sec geometric mean, difference map. Warm colors are higher Study 15.12.
Study 15.12 vs Study 15.4, 10 sec geometric mean, difference map. Warm colors are higher Study 15.12.
Ratio
Study 15.12 vs Study 15.4, 2 sec geometric mean, ratio map. Warm colors are higher Study 15.12.
Study 15.12 vs Study 15.4, 3 sec geometric mean, ratio map. Warm colors are higher Study 15.12.
Study 15.12 vs Study 15.4, 5 sec geometric mean, ratio map. Warm colors are higher Study 15.12.
Study 15.12 vs Study 15.4, 10 sec geometric mean, ratio map. Warm colors are higher Study 15.12.

Comparison with NGA results

We would expect the stochastic results to show reasonably good agreement with the deterministic results, but they do not. Here are some data products to investigate this. We've chosen 3 sites with large differences (MRVY, s028, LAPD) and 2 we're familiar with (PAS, USC).

NGA-2 curve comparisons

We replotted the curves using NGA-2s (instead of NGA-1) for comparison, and RotD50.

Site 0.5 sec 0.2 sec
USC
PAS
MRVY
s028
LAPD

Changing RVSIG

We reran PAS using rvsig1 = 0.1; comparison curves are below.

Blue: rvsig1=0.1, black: rvisg1=-1.0
Blue: rvsig1=0.1, black: rvisg1=-1.0

Disaggregations

We did disaggregations for CyberShake and the 4 NGA-2s at 0.5 sec and 0.2 sec, at prob=1e-4. Below are images for each, along with the top 5 contributing sources:

Data Source 0.5 sec value at prob=4e-4 0.5 sec plot 0.5 sec contributors 0.2 sec value at prob=4e-4 0.2 sec plot 0.2 sec contributors
CyberShake 1.08g
47.0% Verdugo (source 283) 
 23.4% Elysian Park (Upper) (source 158)
 12.5% Raymond (source 247)
 10.6% Puente Hills (LA) (source 244)
  5.9% Puente Hills (source 242)
ASK2014 1.24g
BSSA2014 1.51g
CB2014 1.59g
CY2014 1.65g
Retrieved from "https://strike.scec.org/index.php?title=Study_15.12_Data_Products&oldid=16554"