
A statistical method that averages data over multiple occurrences of a specific event to isolate its common signal from background variability.Composite analysis is a way to find a clear pattern hidden in noisy data by lining up many separate instances of the same kind of event and averaging them. For example, if you want to know what the weather typically does after a volcanic eruption, you take all the eruptions on record, align them at the moment of eruption, and average the temperature readings for each day before and after. Random fluctuations—like a stray heat wave or a passing storm—tend to cancel out in the average, while any consistent response to the eruption stands out clearly.
The technique has roots in early 20th-century meteorology, where it was used to study the effects of phenomena like solar flares on weather, but its formalization as a standard tool in climate science came later.
The method works by selecting a set of time windows, each centered on an occurrence of the event of interest. For each time offset relative to the event, you compute the mean of the variable across all windows. The result is a single averaged curve that represents the typical evolution around the event. Statistical tests, such as comparing the average to the variability across windows, help decide whether the apparent signal is real or just noise.
It is widely used in climate research to isolate the impacts of events like El Niño, volcanic eruptions, or sudden stratospheric warmings from the background of natural variability.
Composite analysis is sometimes called 'superposed epoch analysis,' a name that reflects its use in astronomy to study the effects of solar storms on cosmic ray counts.





