Abstract
Polynomial interpolation and Holt-Winters exponential smoothing (HWES) are used to analyze and forecast Caspian Sea level anomalies derived from 15-year Topex/Poseidon (T/P) and Jason-1 (J-1) altimetry covering 1993 to 2008. Because along-track altimetric products may contain temporal and spatial data gaps, a least squares polynomial interpolation is performed to fill the gaps of along-track sea surface heights used. The modeling results of a 3-year forecasting time span (2005 - 2008) derived using HWES agree well with the observed time series with a correlation coefficient of 0.86. Finally, the 3-year forecasted Caspian Sea level anomalies are compared with those obtained using an artificial neural network method with reasonable agreement found.
Original language | English |
---|---|
Pages (from-to) | 521-530 |
Number of pages | 10 |
Journal | Terrestrial, Atmospheric and Oceanic Sciences |
Volume | 24 |
Issue number | 4 PART1 |
DOIs | |
Publication status | Published - 2013 Aug 1 |
All Science Journal Classification (ASJC) codes
- Oceanography
- Atmospheric Science
- Earth and Planetary Sciences (miscellaneous)