| Sea-level is a fundamental climate parameter, and of particular interest is the description and quantification of long-term sea-level variability. Sea-level time series over a global regular grid are available from both model runs and satellite altimetry observations. However, the analysis of long-term variability requires approaches aimed to efficiently extract the long-term patterns from huge spatio-temporal datasets while allowing to isolate the influence of phenomena such as ENSO on the derived patterns. In this study a new method, trend empirical orthogonal function analysis (trend-EOF), is applied for extracting long-term space-time patterns of sea-level variability from satellite altimetry and model data. The approach is particularly adequate for extracting trend patterns from short time series, such as the ones resulting from satellite retrievals, and is applied here to the global dataset comprising almost 17 years of satellite altimetry observations and to the outputs from model runs. The resulting modes allow to characterise spatial trends in sea-level and to isolate the influence of ENSO in global sea-level variability. |
|