![]() ![]() (right) The autocorrelation of the Niño-3.4 index for 1–6-yr lags is plotted, with the red line indicating significance at the 95% level (2 times the large-lag standard error). (left) Correlation between the DJF Niño-3.4 index and the DJF SST field for each model and the observed SST dataset. The bottom panel shows the range in the sum of squared correlation coefficients over North America for the model segments with the value from the observations and observation-to-reconstruction record plotted as dashed gray and black lines, respectively. ![]() The middle panel shows the range in CPCS between the model pattern for each 152-yr segment and the observation-to-reconstruction pattern. The plotted pattern is the 152-yr segment with the teleconnection pattern that best matches the observed–reconstruction pattern as determined by the CPCS between the two fields. For the models, the teleconnection pattern was calculated for a sliding 152-yr window (the length of the observed record). The observed–reconstruction correlations (RECON) are from the overlapping period (1854–2005) between the NADA and the observed SST dataset ( Smith and Reynolds 2003) and the observed correlations (OBS) are from the 1950–2005 period in an observed PDSI dataset ( Dai 2004) and the same SST dataset. For the PDO and AMO, correlations are based on filtered PDSI and climate indices. ![]() Correlations between gridpoint PDSI from the NADA or models and the (top) DJF Niño-3.4 index, (middle) DJF PDO index, and (bottom) JJA AMO index.
0 Comments
Leave a Reply. |