Global warming is a well-known and well-studied phenomenon pertaining to a gradual increase of average global temperatures over time. Many global warming mathematical models make certain assumptions regarding the factors that impact global temperature. These assumptions include effects from increased global carbon dioxide levels in the atmosphere and the melting ice sheets, among others. This paper draws conclusions about temperature changes without the assumptions needed for the global warming mathematical models. Instead of using computer models to project temperatures on a global scale, 33 low-latitude locations in the southern United States were individually studied to see if each one has warmed over time. Each location's daily high temperature was obtained for each day since January 1, 1970, and the data is analyzed using a statistical model that contains a linear effect coefficient, an annual seasonal trend, and a 10.7-year solar cycle trend, both well-known and commonly-accepted periodic trends. The data is also analyzed using Fourier frequencies to check for other lesser-known periodic trends. The strongest trend is then added to the original model. Linear effect coefficients are calculated for each location using both the updated model and the original model to see how they compare to each other and global warming study results. Only using the 31 locations where these models fit, the original model yielded an average linear increase of 2.29 degrees Celsius per century, while the updated model showed an average linear increase of 2.33 degrees Celsius per century. Showing less than a 2% difference indicates that the original model is sufficient and that temperatures in low-latitude locations are increasing.
David Prager, Assistant Professor of Mathematics, Anderson University
"Local Warming in Low Latitude Locations: A Time Series Analysis,"
Rose-Hulman Undergraduate Mathematics Journal: Vol. 19
, Article 7.
Available at: https://scholar.rose-hulman.edu/rhumj/vol19/iss1/7