Climate change assessment using statistical process control methods

Abstract

Statistical process control (SPC) uses the application of statistical methods and procedures to monitor and control a process, in order to evaluate two possible causes of variation in the process: natural (common) and assignable (special) causes. The aim of this activity is to improve the process’ capabilities. If the variability of a process is within the range of natural causes, the process is said to be under statistical control. When that variability exceeds the expected natural causes range, it is a signal to look for, and to correct, assignable causes. SPC may even be used to “control” climate change, through comparison of present day variations with the natural variation capacity for change in air temperature, precipitation and sea levels in the past. Are today’s frequent floods, tornados, warm winter periods or cold summer days actually caused by “natural” causes (should they be statistically expected), or has the “capability” of the natural processes changed? This paper will demonstrate the potential use of SPC methods in evaluating variations in temperature and precipitation that should be expected, based on the assessment of the statistical behaviour of data for these natural indicators during different periods. “Warning” and “action” lines will be assessed and compared for the selected periods. Also, the number of records below or above warning and action lines will be compared. This approach could be useful for spatial planners, even if the causes of the changes are global or not human-induced.