Metr 485/785 Homework 3

Reproducability of results: an exercise in application of the scientific method: Part I

You have read the Ladochy et al. paper. The purpose of this exercise is to attempt to duplicate some of the results that the authors summarize with respect to the relationship of PDO, SOI and NINO3 indices to rainfall in California. You will note that the author placed his email and mailing addresses on the first page of the manuscript. This is to allow for scholarly dialog about the results and procedures summarized in the manuscript.

The authors and this work were widely quoted in 2004, particularly the contention that PDO is decreasing, and that rainfall in Southern California had been decreasing will be expected to continue to decrease in the years following. A part of the scientific method involves results being reproducible. We will perform some simple statistical analyses to verifty the authors' contentions.

The purpose of this exercise is not to encourage carping or unbridled skepticism without reason. The purpose is to have you prove to yourself the degree to which the results summarized in the paper can be reproduced or are statistically valid, either in support of or in disagreement with the claims of the authors.

Answer the following questions:

1. (Due Friday 22 February) Is the material in the Data and Methodology Section consistent with what a reader would want to know about what was done in the paper with respect to the relationship of the various indices LaDochy studies and precipitation in Central and Southern California (in other words, are all sources of data and methodologies outlined and summarized? Discuss.
   
   
2. (Due Friday 22 February, first pass) Using the Excel Spreadsheet of LA Civic Center Rainfall create a graph of seasonal rainfall overlain with 25 yr running mean of seasonal rainfall,
   
   
3. (Due Friday 22 February, first pass) Using the Excel Spreadsheet of LA Civic Center Rainfall and PDO values, produce the following: (a) Graph of 25 yr running mean of Coefficient of Variation 25 yr running mean of seasonal rainfall; (b) Graph of 25 yr running mean of seasonal rainfall and 25 yr running mean of PDO.
   (Note: to produce the 25 yr running mean of coefficient of variation, you will not be able to use Excel's built in "moving average" trendline feature. That is because the coefficient of variation itself is based upon the standard deviation that is in turn calculated on the basis of data evaluated over a long term. To do this, calculate the 25 yr coefficient variation for the period ending 1946 (that will be 25 years). Then copy this formula and paste it into the cells underneath for the remainder of the period of record, remembering not to include 2001 since the rainfall data is incomplete for that season). Do this for both the seasonal rainfall and the PDO indices to get the graphs above).