This directory contains extra materials relevant for the Schmidt, Mann and
Rutherford comment on the McShane and Wyner (2010, AOAS) paper. Please
send all comments or corrections to Gavin Schmidt <gschmidt@giss.nasa.gov>.

The EIV/ directory contains a cleaned up version of code originally
archived for Mann et al (2007, JGR) here: 

http://www.meteo.psu.edu/~mann/PseudoproxyJGR06/

(including later corrections). 

The EIV/code/ directory contains a turnkey MATLAB script

   master_pseudo_eiv.m

which uses the same pseudo-proxies and temperature targets as were
used for the M&W methods in the original supplemental material. The
resulting reconstructions are slightly different from those originally
archived due to a) slightly different target temperatures, b) an
inadvertent hand coding error in defining regpar in one nohybrid case,
and c) they are annual reconstructions (instead of decadally smoothed
ones which were mistakenly archived orginally). The differences in
these reconstructions compared to the original ones can be seen most
clearly in SMR_figextra1.pdf.

The routine "eiv_recon.m" is a single function call that optionally
produces the hybrid or non-hybrid reconstruction for each target and set
of pseudo-proxies. The original targets and pseudo-proxies are
duplicated in the EIV/data/pseudo/ directory.

Output is in the EIV/output/ directory and consists of the annual
reconstructions as well as the 10yrsmooth reconstructions for easy
comparison to the previous versions (which are in EIV/orig_output).

There are three R scripts: 

R_figextra1.txt plots the comparisons of the new and original reconstructions.  

R_fig2update.txt replots fig 2 from the SMR comment with the more
consistent EIV reconstructions, and produces the Table of skill scores
that are defined using a common baseline of 1856-1980 for all
reconstrcution methods.

R_pseudo_v2.txt is an updated version on R_pseduo.txt from the
original archive.