Identifying Sudden Changes in Data

J. F. Howell Oceanic and Atmospheric Sciences, Oregon State University, Corvallis, Oregon

Search for other papers by J. F. Howell in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

This note describes a method for locating sudden changes in mean data values. Positions of sudden changes are boundaries of variable-width blocks of data. These boundaries could correspond to synoptic frontal boundaries, the downstream edge of a wind gust, or, generally, any anomalous change in a locally averaged quantity. The algorithm described here is applied to artificial signals, century-long records of precipitation, and atmospheric turbulence data.

Abstract

This note describes a method for locating sudden changes in mean data values. Positions of sudden changes are boundaries of variable-width blocks of data. These boundaries could correspond to synoptic frontal boundaries, the downstream edge of a wind gust, or, generally, any anomalous change in a locally averaged quantity. The algorithm described here is applied to artificial signals, century-long records of precipitation, and atmospheric turbulence data.

Save