In the natural sciences, especially in atmospheric and Earth sciences involving applied statistics, an anomaly is a persisting deviation in a physical quantity from its expected value, e.g., the systematic difference between a measurement and a trend or a model prediction.[1]
Similarly, a standardized anomaly equals an anomaly divided by a standard deviation.[1]
A group of anomalies can be analyzed spatially, as a map, or temporally, as a time series.
It should not be confused for an isolated outlier.
There are examples in atmospheric sciences and in geophysics.
The location and scale measures used in forming an anomaly time-series may either be constant or may themselves be a time series or a map. For example, if the original time series consisted of daily mean temperatures, the effect of seasonal cycles might be removed using a deseasonalization filter.
In the atmospheric sciences, the climatological annual cycle is often used as the expected value. Famous atmospheric anomalies are for instance the Southern Oscillation index (SOI) and the North Atlantic oscillation index. SOI is the atmospheric component of El Niño, while NAO plays an important role for Europeanweather by modification of the exit of the Atlantic storm track.
A climate normal can also be used to derive a climate anomaly.[2]
