Reconcile fortran grid, sub, and fault source files

[pmpowers-usgs]

Since the 2014 release, there have been numerous updates to the fortran source codes. These have mostly been in support of adding addition periods via coefficient interpolation and to support induced seismicity.

[pmpowers-usgs]

Header comments added to hazgridxnga13l.f since last update

• nov 2,2015: repair a little bug with "oktogo" variable so that 2-s SA can be computed with Frankel and others
• nov 5, 2015: repair initializations for "getToro" so that 2-s SA works with this GMM. In Toro, the period index for 2-s SA is 7.
• Oct 20, include the Atkinson_2015 Induced seismicity GMM "a15ind" with index 51.
• Nov 18, 2015. Use dtor(1) to model source depth in A15ind. Previously, set at 5 km.
• Aug 31 2015: include the Hawaii terms to BA08 GMM ( use index 50 for Hawaii GMM.) change the unit=15 to unit=95 because of a conflict with station file open.
• Crustal-source Zhao now allowed June 24 2015. Use index 49 for Zhao GMM (29 is BSSA)
• June 2015: include 0.4s for ABslab (getABsub) calcs.
• May 27 2015: changed a few coeffs for soil response in ASK2013 to agree with Spectra and Peer reports
• May 15 2015: add 2.5s coeffs to ASK2013, CB13, CY2013, Idriss13, and AB03(slab)
• April 7, 2015: Correct c(0.4s) in Zhao et al. to 0.01 as in Table 4. Previous value was from an email from Zhao. His email set c(0.4s) to 0.006.
• Feb 13, 2015: Define a clamp for the 14 periods in the three newer Atkinson&Pezeshk CEUS GMMs. Remove the "iq" index from these subroutines. Note: Units of PGV are cm/s in the Atkinson CEUS GMMs. Provisional CEUS PGV clamp is set at 400 cm/s. This value should be discussed.
• Feb 10, 2015: Use a coarser test to accept 3.33 hz as equiv. to 0.3-s in newer CEUS Atkinson GMMs.

[pmpowers-usgs]

Notes on hazFXnga13l.f

• Fixes to reading Gail Atkinsons table files.
• Support for PGV in Gail's tables
• Added support for 1.5 sec in CEUS GMMs via coefficient interpolation

[pmpowers-usgs]

Notes on hazFXnga13p.f

The parallelized version of hazFX* in the repository is well out of sync with the single-threaded version. However, the version supplied by @ashumway-usgs appears to have been derived from her single-threaded version.

[pmpowers-usgs]

Notes on hazSUBXnga.test.f

• Apr 7 2015: The c(0.4s) coef. for Zhao et al. was changed to 0.01 (as in BSSA Table 4) instead of 0.006, from Zhao's spreadsheet. The effect is to remove a spiky amplification at 0.4 s from the Zhao interface spectrum.
• Apr 2 2015: reevaluated 0.15-s, 0.30s &0.75-s coeffs for AM09. Previous version (Dec 3 14) was less accurate.
• Mar 11 2015: Always taper GR distribution above M9.5. Taper rate is sqrt(e) per 0.1 M unit. Taper starts at first sampled M above 9.5
• Dec 3 2014: Add 0.15s to AM09
• June 20 2014: Add 0.75-s coeffs. to AM09. BAsiteamp already has 0.75 coefs.
• Mar 21 2014: add 2.5s to BAsiteamp to use with AM09 if requested. SH.
• Feb 24 2014: Include option to use Papazachos (2004) orStrasser et al (2010) L(M) length as fcn of M as alternatives to the older Geomatrix relation (1995). Papazachos : use itype = 4; Strasser: use itype = 10

[pmpowers-usgs]

I have a local, intermediate file that includes interpolated NGAW2 coefficients for 2.5s