New features of land and sea surface temperature anomalies

Global mean land and sea surface temperature (LSST) anomalies were analysed statistically with detrended fluctuation analysis (DFA), for the period from January 1850 to August 2008, for both hemispheres, individually. Persistent long-range power-law correlations were revealed within the time series of the LSST anomalies, for all time lags between 4 months and 39 years. In addition, the derived 'long memory' in the time series of the LSST anomalies was examined to determine whether they are derived from their time evolution or from the distribution of their own values. The results obtained could be taken into consideration for the prediction of basic climate variables through the use of future climate modelling.