Trends in atmospheric carbon dioxide over the last ten to fifteen years

This study looks at the relative importance of the factors which control the concentration of atmospheric carbon dioxide. EOF analyses are run for both the seasonal and non-seasonal variations for the ten years from 1981 to 1990. The first seasonal EOF represents the anthropogenic component as well as the breathing of the land biosphere. Representing 85% of the variation, it has a seasonal variation of almost 6. The second shows the component of just the land biosphere and has a seasonal variation of about 4. The third seasonal EOF is thought to portray the effect of upwelling in the Eastern Pacific on the tropical strip. The first non-seasonal EOF, accounting for 97% of the total variance, shows an increase of about 11.7 for the period; that of the Northern Hemisphere is about 1.5 times that of the Southern Hemisphere. A modification was made to the original anomalies to adjust for the long term trend in the carbon dioxide data. The rest of the procedure was the same. The first seasonal EOF, contributing to 65.8% of the variance, appears to represent changes in the NH terrestrial biosphere. The second seasonal EOF shows the variance between the long-term trend and the actual data. The third seasonal EOF, with a variance of 4, again depicts the oceanic variance due to upwelling off the coast of Peru. The first non-seasonal EOF matches the activity of the El Nifio, supporting the theory that upwelling increases atmospheric CO2 concentration. It represents over 50% of the total variation. Error in this study may stem from unreliable station data due to infrequent sampling, not a long enough time period for the analysis, and not enough stations to develop a good global result. A stronger, standardized network would greatly enhance the outcome of this analysis. Thesis Supervisor: Reginald Newell Title: Professor Acknowledgments I would like to thank Professor Reginald Newell for all his help and understanding as my advisor. He always managed to find that one piece of information I was looking for, and also helped me through some difficult times. To Wenjie Hu, I am forever indebted for her help with my attempts to program in a language which I knew nothing about, not to mention teaching me everything she new about eigenvector analysis. No matter how matter how many times I asked, she patiently explained everything until I finally understood it. I also want to thank my parents, Ronald and Arlene Strader, for their help and support of me in everything, no matter how small, that I have ever tried to do. They have always been there when I needed them. To my best friends Michelle Bakkila, John Hansen, Andrea Jensen, and Theresa Hutchings, I just want to say, I don't know how I could have gotten through and remained sane without you.