An Investigation for the Trend of Global Warming Based on the Global Seawater Oxygen-18 Database from NASA

• Activity Number: 59
• Type: Contributed
• Date/Time: Sunday, August 3, 2008 : 4:00 PM to 5:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Qiqi Deng*+ and Snigdhansu Chatterjee
• Companies: The University of Minnesota and The University of Minnesota
• Keywords: Polynomial local regression ; CUSUM

Abstract

Background: NASA Goddard Institute for Space Studies (GISS) has the Global Seawater Oxygen-18 Database with over 22,000 observations recorded since about 1950, including date, ocean temperature, depth and coordinates, etc. In this study, we use this database to investigate the trend of global warming. Method: Polynomial local regression is applied to the data after appropriate transformations. Residual plots are checked for model validation. CUSUM chart and other statistical methods are used to detect the change of ocean temperature along years after adjusting the influences by other predictors, such as depth, spatial effects and seasonal effects. Results: The relationship between the ocean temperature and the predictors are complicated, and some interactions exist. The CUSUM chart suggests that there is an indication of slightly increasing trend for ocean temperature over years.