2001•Geophysical Research LettersOpen access
Singular spectrum analysis for time series with missing data
David H. Schoellhamer
Open full text 220 citations
Abstract
Geophysical time series often contain missing data, which prevents analysis with many signal processing and multivariate tools. A modification of singular spectrum analysis for time series with missing data is developed and successfully tested with synthetic and actual incomplete time series of suspended‐sediment concentration from San Francisco Bay. This method also can be used to low pass filter incomplete time series.
Open-access reader