# -----------------------------------------------------------------------------
# Automated Phanerozoic Biodiversity Curve Generator
# Bjorn-Gustaf J. Brooks, November 8, 2007
# -----------------------------------------------------------------------------

In order to describe the methods necessary to replicating my Phanerozoic
biodiversity curves using data downloaded from the Paleobiology Database
(www.paleodb.org), I have created a simple BASH script executable that
makes calls to other AWK, BASH, C-shell, PSQL, and SED scripts in order to
automatically generate PostgreSQL tables and query those tables to provide
the necessary interpolated Phanerozoic biodiversity counts sampled at
1 Myr increments. These programs are standard in most or all Linux
distributions. If you are running Linux, make sure you have PSQL
priveledges to create databases and tables.

Once you have extracted the tar file, 'cd' into the 'supplement' directory.
Next execute the create_db.bash script.


# -----------------------------------------------------------------------------
# GENERATE PHANEROZOIC BIODIVERSITY DATA --------------------------------------
If you are running PSQL and wish to generate our summary age-range data
tables and biodiversity query results execute the following command:

   ./create_db.bash

This will take several hours to populate your database, find the first
and last occurrences (FADs & LADs) of each genus name, generate the
interpolated age ranges for each genus name and plot your graphs with
the Monte Carlo trials. Determining the FADs & LADs is the step that takes
the longest, by far. You may want to use nohup and dump standard error from
/dev/null into a log file, so that you can log out, and have a hot cup
of coffee while the machine toils away for you:

   (nohup time ./create_db.bash < /dev/null ) >& create_db.log &

After the script is complete the results to your query will be output as .csv
files in the various subdirectories of 'query' such as 'deep', 'marine', and
'marine_indet'. These two column .csv files contain the time and biodiversity
counts necessary for your spectral analysis.


# -----------------------------------------------------------------------------
# GENERATE PHANEROZOIC BIODIVERSITY DATA & PLOTS ------------------------------
If you are also running Octave or Matlab and wish to generate our plots, which 
will also plot the average of the Monte Carlo random step and random block 
trial routines. Remember that to save time I have preset the Monte Carlo loop 
to 200 trials, not the 10,000 trials which we used in our paper. You will have 
to navigate to the directory of interest (supplement/g21663/query/gba/) and 
edit the Matlab/Ocatave file 'dft.m', substituting 'numtest=200' with 
'numtest=10000' in both places within the file (one place for the number of 
random step iterations and the second for the number of random block 
iterations). 


# -----------------------------------------------------------------------------
# WHERE TO FIND OUR PAPER -----------------------------------------------------

    http://www.climatemodeling.org/paleo/diversity/

# -----------------------------------------------------------------------------
# WHERE TO FIND OUR ROUTINES --------------------------------------------------

    http://www.climatemodeling.org/paleo/diversity/supplement.tar.gz
    http://www.climatemodeling.org/paleo/diversity/supplement.zip

# -----------------------------------------------------------------------------
# DOCUMENTATION ---------------------------------------------------------------
These methods are described in detail through our Supplementary Information
of 'Sources of Cycles in Phanerozoic Biodiversity', and also here:

    http://www.climatemodeling.org/paleo/diversity/

Enjoy!