[cig-commits] r13471 - in seismo/3D/ADJOINT_TOMO/flexwin/latex: . figures figures/reviewer_response

[alessia at geodynamics.org]

Author: alessia
Date: 2008-12-08 04:22:56 -0800 (Mon, 08 Dec 2008)
New Revision: 13471

Added:
   seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/
   seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV.IU.LHZ.seis_stalta.pdf
   seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV_IU_LHZ_seis_stalta.pdf
Modified:
   seismo/3D/ADJOINT_TOMO/flexwin/latex/response.pdf
   seismo/3D/ADJOINT_TOMO/flexwin/latex/response.tex
Log:
Finished response letter.

Added: seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV.IU.LHZ.seis_stalta.pdf
===================================================================
(Binary files differ)


Property changes on: seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV.IU.LHZ.seis_stalta.pdf
___________________________________________________________________
Name: svn:mime-type
   + application/octet-stream

Added: seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV_IU_LHZ_seis_stalta.pdf
===================================================================
--- seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV_IU_LHZ_seis_stalta.pdf	                        (rev 0)
+++ seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV_IU_LHZ_seis_stalta.pdf	2008-12-08 12:22:56 UTC (rev 13471)
@@ -0,0 +1 @@
+link OTAV.IU.LHZ.seis_stalta.pdf
\ No newline at end of file


Property changes on: seismo/3D/ADJOINT_TOMO/flexwin/latex/figures/reviewer_response/OTAV_IU_LHZ_seis_stalta.pdf
___________________________________________________________________
Name: svn:special
   + *

Modified: seismo/3D/ADJOINT_TOMO/flexwin/latex/response.pdf
===================================================================
(Binary files differ)

Modified: seismo/3D/ADJOINT_TOMO/flexwin/latex/response.tex
===================================================================
--- seismo/3D/ADJOINT_TOMO/flexwin/latex/response.tex	2008-12-08 11:01:16 UTC (rev 13470)
+++ seismo/3D/ADJOINT_TOMO/flexwin/latex/response.tex	2008-12-08 12:22:56 UTC (rev 13471)
@@ -1,6 +1,7 @@
 \documentclass{letter}
 \usepackage{a4wide}
 \usepackage{setspace}
+\usepackage{graphicx}
 \onehalfspacing
 \renewcommand{\deg}{$^\circ$}
 %\begin{document}
@@ -31,7 +32,7 @@
 In the following, we reply in detail to the major comments made by the
 reviewers.  Quoted comments are in italics.
 
-Reviewer 1 (Anonymous):
+Reviewer 1 :
 
 
 \begin{enumerate}
@@ -138,7 +139,7 @@
 arrivals generated by fully 3D strucutres, for which attempting to fit named phases
 would not be productive.
 %
-\item{\em Incomplete list of references: although itself a data processing
+\item{\label{REF}\em Incomplete list of references: although itself a data processing
 paper, the present work does not put itself in the context of other recent work
 on measurements and data processing for waveform-based tomography.  Earlier
 work may not be geared specifically at adjoint inversions, but the challenge of
@@ -165,11 +166,86 @@
 (page 6, last paragraph).
 \end{enumerate}
 
+\newpage
+Reviewer 2 :
 
+\begin{enumerate}
+\item{\em Automated analysis has been done for quite a while now, both with a
+focus on surface waves and body waves.  Some review of this work would be
+useful, particularly since some of the quality parameters that Maggi et al.
+invoke are similar to those used before.}
+\\
+See response to point~\ref{REF} above.
+\item{\em The noise criterion of Phase 0 should be set and defined very
+carefully.  The authors show a pretty nice example of a nicely excited Pdiff
+that appears to be rejected from further analysis given its relatively low SNR.
+Pdiff are extremely important phases.  Without them, you cannot image the base
+of the mantle.}
+\\
+In the revised manuscript, Phases 0 and A-D have been renamed as Stages A-E.
+The noise criterion of Phase 0 thus becomes the noise criterion of Stage A.
+This criterion (actually two criteria, based on the amplitude and power of the
+signal and noise) determines the rejection of the observed seismogram as a
+whole.  As it seems there was some confusion on this point, it has been
+emphasized in Appendix A.  The SNR criterion that causes the Pdiff to be
+rejected is the window signal to noise ratio $r_0(t)$ of Stage D (formerly
+Phase C).  This criterion is
+time-dependent, and can therefore be modulated to favour the acceptance of
+important, low-amplitude phases such as Pdiff.  An indication of how this could
+be achieved is given in Appendix A.
+\item{\em There is no indication on how well FLEXWIN will perform for smaller
+(more frequent) earthquakes.  A general discussion on FLEXWIN performance for
+noisier data would be useful.}
+\\
+We agree that an example showing how FLEXWIN performs for smaller events in our
+long-period global tomography example was missing from the original manuscript.
+We have therefore added a fourth event to those discussed in Section 3 (a M5.7 in the
+Comoros Region, see Figure 7c,d).  The smaller the event, the more prominent
+the long-period noise.  Therefore in order to use events smaller than this, it
+would be necessary to analyse shorter periods.  
+\item{\em For smaller earthquakes it is probably necessary to analyze signals
+with periods much shorter than 50s.  How well does FLEXWIN handle itself with
+S20 and crust2.0 at, say, 20s?}
+\\
+We agree entirely that for smaller earthquakes it would be necessary to analyse
+shorter period signals.  In order to keep the paper as simple as
+possible, we have chosen to limit discussion of global tomography to the long
+period case, for which (as dicussed above) smaller earthquakes do not generate
+much useable signal.  In response to the question posed, FLEXWIN handles itself
+well at 20 seconds with synthetics created using S20RTS and CRUST2.0.
+The figure below shows the FLEXWIN results for the same path
+as that shown in Figure 7c of the revised manuscript, analysed between 20 and
+150 seconds period.  All parameters other than $T_0$ were kept unchanged for
+this run.  Even without specific tuning, the windows chosen by the algorithm
+for these shorter-period signals are more than reasonable.
+{\center \includegraphics[width=5in]{figures/reviewer_response/OTAV_IU_LHZ_seis_stalta.pdf}}
+\item{\em I cannot follow the discussion at the bottom of page 13 and top of
+page 14.  I am not sure which of the signals is supposedly "not a seismic phase
+in the high-frequency sense... but undoubtedly contains information".}
+\\
+We have clarified this point in the revised manuscript (page 15, second
+paragraph) as follows:  Not all features within a given seismogram are
+identifiable as seismic phases.  For example, the second window in Figure 8b
+seems to contain two features.  When we look at periods shorter than 50s, the
+first feature retains it character and is clearly identifiable as sSdiff, while
+the second feature looses its character entirely and is more readily
+assimilated to a generic S wave coda than to a distinct seismic phase.  This
+feature is present in both observed and synthetic sismograms, and undoubtedly
+contains information.  
+\item{\em I think SPECFEM uses the depth dependent scaling that came out of Ritsema
+and van Heijst, GJI, 2002.  Perhaps the used even the 3D P12 model.  In any
+case, it might be useful to emphasize that the scaling involves more than just
+a scale factor.}
+\\
+We thank the reviewer for raising this point.  Indeed, the SPECFEM
+implementation used to generate the synthetic seismograms uses the P12 model of
+Ritsema and van Heijst, GJI, 2002.  We have modified the text accordingly (page
+14, section 3.1, paragraph 1).
+\end{enumerate}
 
+We trust this revised manuscript will meet with your approval.  
 
 
-
 \closing{Sincerely yours,}