论文部分内容阅读
本文利用大范围月表面水温资料,经过空间平均,减去多年月平均后,获得了热带东太平洋冷水舌域的表面水温距平逐月时间序列(1951年1月—1985年12月).并依此分别建立了自回归模式,自激励门限自回归模式和开环门限自回归模式.应用这三种模式对上述区域的水温距平年际变化曲线进行了模拟.结果表明,所有三个模式对1951年以来的9次埃尔尼诺过程均做出了很好的后报.为进一步验证模式的可靠性,特将该模式进行了外推预报检验(外推时间为1986年1月—1987年2月).预报结果表明,该模式较好地预测出最近的埃尔尼诺过程(1986—1987)的发生和发展强度.三种模式的模拟和预测结果与实测值的相关系数分别是0.84,0.88和0.89.可见三种模式均好,尤以开环门限自回归模式为最佳.这说明开环门限自回归模式是一个监测东太平洋冷水舌域水温距平的年际变化和估算埃尔尼诺强度的有力工具。
In this paper, monthly surface water temperature anomalies over the cold water in the tropical Pacific Ocean are obtained from the monthly average of the monthly mean surface water temperature after deducting the monthly average over many years (January 1951 - December 1985) The autoregressive threshold, the self-excitation threshold autoregressive and the open-loop threshold autoregressive are respectively established, and the interannual variation curves of water temperature and temperature are simulated using the three models.The results show that all three modes Nine El Niño processes have been well reported since 1951. In order to further verify the reliability of the model, the model was extrapolated and tested (extrapolated from January 1986 to 1987 February) .The forecast results show that this model predicts the occurrence and intensity of the recent El Niño process (1986-1987) .The correlation coefficients between the simulated and predicted results of the three models and the measured values are 0.84, 0.88 and 0.89, showing that all three models are good, especially the open-loop threshold autoregressive model is the best, which shows that the open-loop threshold autoregressive model is an interannual model to monitor the interannual variation of water temperature anomalies in the cold water tongue of the eastern Pacific Ocean El Nino strength A powerful tool.