利用IERS(International Earth Rotation and Reference Systems Service)发布的EOP 14 C04产品中1984~2022年LOD实测值进行频谱分析及周期项提取,并采用最小二乘外推模型联合多项式曲线拟合模型方法将提取到的周期项应用于对LOD序列的...利用IERS(International Earth Rotation and Reference Systems Service)发布的EOP 14 C04产品中1984~2022年LOD实测值进行频谱分析及周期项提取,并采用最小二乘外推模型联合多项式曲线拟合模型方法将提取到的周期项应用于对LOD序列的拟合。实验结果表明,相比于单一的LS外推模型拟合,新方法拟合序列的RMSE从0.0003 s下降至0.0001 s,且新方法将确定系数从0.80提高到0.97左右。该研究结果可为LOD序列的预报研究提供参考。展开更多
In this paper the multi-stage digit filter is used to analyse the data of Earth rotation represented by the length of day, ΔLOD. The results show that the interannual variations of Earth rotation, which are in the ti...In this paper the multi-stage digit filter is used to analyse the data of Earth rotation represented by the length of day, ΔLOD. The results show that the interannual variations of Earth rotation, which are in the time scale of several years but not quasi-periodic terms, exist in the long periodic fluctuations. They induce the relative variation in the length of day of 0.3×10^(-8).Comparing the series of length of day with the data of temperature departure of the sea surface in the equatorial area of the eastern Pacific, we found that the deceleration and acceleration of the interannual rate of Earth rotation are consistent with the warming up and down of sea temperature in the equatorial area very well. This means that every El Nio event always occurs after the turning of acceleration of the interannual rate of Earth rotation to deceleration.According to the strong interannual variation in the length of day and strong warming of the sea surface temperature in the equatorial area between 1982 and 1983, we analysed the data from atmospheric angular momentum (AAM) calculated by using the global zonal wind data, and found that the interannual variation in AAM has an excess of two to three months. We suggest that the interannual variations in Earth rotation and the El Nio events are probably responses of solid earth and ocean, respectively, to the anomaly of atmospheric circulation.It is also shown in oar analysis that the minimum of ΔLOD series, which is deduced from UT1 data observed regularly with astrometry, can predict the occurrence of the El Nio events for a long range forecast about one year.展开更多
Variation in length of day of the Earth （LOD, equivalent to the Earth＇s rotation rate） versus change in atmospheric geopotential height fields and astronomical parameters were analyzed for the years 1962-2006. This...Variation in length of day of the Earth （LOD, equivalent to the Earth＇s rotation rate） versus change in atmospheric geopotential height fields and astronomical parameters were analyzed for the years 1962-2006. This revealed that there is a 27.3-day and an average 13.6-day periodic oscillation in LOD and atmospheric pressure fields following lunar revolution around the Earth. Accompanying the alternating change in celestial gravitation forcing on the Earth and its atmosphere, the Earth＇s LOD changes from minimum to maximum, then to minimum, and the atmospheric geopotential height fields in the tropics oscillate from low to high, then to low. The 27.3-day and average 13.6-day periodic atmospheric oscillation in the tropics is proposed to be a type of strong atmospheric tide, excited by celestial gravitation forcing. A formula for a Tidal Index was derived to estimate the strength of the celestial gravitation forcing, and a high degree of correlation was found between the Tidal Index determined by astronomical parameters, LOD, and atmospheric geopotential height. The reason for the atmospheric tide is periodic departure of the lunar orbit from the celestial equator during lunar revolution around the Earth. The alternating asymmetric change in celestial gravitation forcing on the Earth and its atmosphere produces a ＂modulation＂ to the change in the Earth＇s LOD and atmospheric pressure fields.展开更多
The time-integrated yearly values of North Atlantic Oscillation (INAO) are found to be well correlated to the sea surface temperature. The results give the feasibility of using INAO as a good proxy for climate change ...The time-integrated yearly values of North Atlantic Oscillation (INAO) are found to be well correlated to the sea surface temperature. The results give the feasibility of using INAO as a good proxy for climate change and contribute to a more complete picture of the full range of variability inherent in the climate system. Moreover, the extrapolation in the future of the well identified 65-year harmonic in INAO suggests a gradual decline in global warming starting from 2005.展开更多
针对日长(Length Of Day,LOD)变化预报中最小二乘(Least Squares,LS)拟合存在端点效应的问题,采用时间序列分析方法对日长变化序列进行端点延拓,形成一个新序列,然后用新序列建立最小二乘模型,最后再结合最小二乘模型和自回归(Autoregre...针对日长(Length Of Day,LOD)变化预报中最小二乘(Least Squares,LS)拟合存在端点效应的问题,采用时间序列分析方法对日长变化序列进行端点延拓,形成一个新序列,然后用新序列建立最小二乘模型,最后再结合最小二乘模型和自回归(Autoregressive,AR)模型对原始日长变化序列进行预报。实验结果表明,在日长变化序列两端增加统计延拓数据,能有效减小最小二乘拟合序列的端点畸变,从而提高日长变化的预报精度,尤其对中长期预报精度提高明显。展开更多
Traditional artificial neural networks （ANN） such as back-propagation neural networks （BPNN） provide good predictions of length-of-day （LOD）. However, the determination of network topology is difficult and time ...Traditional artificial neural networks （ANN） such as back-propagation neural networks （BPNN） provide good predictions of length-of-day （LOD）. However, the determination of network topology is difficult and time consuming. Therefore, we propose a new type of neural network, extreme learning machine （ELM）, to improve the efficiency of LOD predictions. Earth orientation parameters （EOP） C04 time-series provides daily values from International Earth Rotation and Reference Systems Service （IERS）, which serves as our database. First, the known predictable effects that can be described by functional models-such as the effects of solid earth, ocean tides, or seasonal atmospheric variations--are removed a priori from the C04 time-series. Only the residuals after the subtraction of a priori model from the observed LOD data （i.e., the irregular and quasi-periodic variations） are employed for training and predictions. The predicted LOD is the sum of a prior extrapolation model and the ELM predictions of the residuals. Different input patterns are discussed and compared to optimize the network solution. The prediction results are analyzed and compared with those obtained by other machine learning-based prediction methods, including BPNN, generalization regression neural networks （GRNN）, and adaptive network-based fuzzy inference systems （ANFIS）. It is shown that while achieving similar prediction accuracy, the developed method uses much less training time than other methods. Furthermore, to conduct a direct comparison with the existing prediction tech- niques, the mean-absolute-error （MAE） from the proposed method is compared with that from the EOP prediction comparison campaign （EOP PCC）. The results indicate that the accuracy of the proposed method is comparable with that of the former techniques. The implementation of the proposed method is simple.展开更多
文摘In this paper the multi-stage digit filter is used to analyse the data of Earth rotation represented by the length of day, ΔLOD. The results show that the interannual variations of Earth rotation, which are in the time scale of several years but not quasi-periodic terms, exist in the long periodic fluctuations. They induce the relative variation in the length of day of 0.3×10^(-8).Comparing the series of length of day with the data of temperature departure of the sea surface in the equatorial area of the eastern Pacific, we found that the deceleration and acceleration of the interannual rate of Earth rotation are consistent with the warming up and down of sea temperature in the equatorial area very well. This means that every El Nio event always occurs after the turning of acceleration of the interannual rate of Earth rotation to deceleration.According to the strong interannual variation in the length of day and strong warming of the sea surface temperature in the equatorial area between 1982 and 1983, we analysed the data from atmospheric angular momentum (AAM) calculated by using the global zonal wind data, and found that the interannual variation in AAM has an excess of two to three months. We suggest that the interannual variations in Earth rotation and the El Nio events are probably responses of solid earth and ocean, respectively, to the anomaly of atmospheric circulation.It is also shown in oar analysis that the minimum of ΔLOD series, which is deduced from UT1 data observed regularly with astrometry, can predict the occurrence of the El Nio events for a long range forecast about one year.
基金supported by the National Science Foundation of China (Grant No 40675031)
文摘Variation in length of day of the Earth （LOD, equivalent to the Earth＇s rotation rate） versus change in atmospheric geopotential height fields and astronomical parameters were analyzed for the years 1962-2006. This revealed that there is a 27.3-day and an average 13.6-day periodic oscillation in LOD and atmospheric pressure fields following lunar revolution around the Earth. Accompanying the alternating change in celestial gravitation forcing on the Earth and its atmosphere, the Earth＇s LOD changes from minimum to maximum, then to minimum, and the atmospheric geopotential height fields in the tropics oscillate from low to high, then to low. The 27.3-day and average 13.6-day periodic atmospheric oscillation in the tropics is proposed to be a type of strong atmospheric tide, excited by celestial gravitation forcing. A formula for a Tidal Index was derived to estimate the strength of the celestial gravitation forcing, and a high degree of correlation was found between the Tidal Index determined by astronomical parameters, LOD, and atmospheric geopotential height. The reason for the atmospheric tide is periodic departure of the lunar orbit from the celestial equator during lunar revolution around the Earth. The alternating asymmetric change in celestial gravitation forcing on the Earth and its atmosphere produces a ＂modulation＂ to the change in the Earth＇s LOD and atmospheric pressure fields.
文摘The time-integrated yearly values of North Atlantic Oscillation (INAO) are found to be well correlated to the sea surface temperature. The results give the feasibility of using INAO as a good proxy for climate change and contribute to a more complete picture of the full range of variability inherent in the climate system. Moreover, the extrapolation in the future of the well identified 65-year harmonic in INAO suggests a gradual decline in global warming starting from 2005.
文摘针对日长(Length Of Day,LOD)变化预报中最小二乘(Least Squares,LS)拟合存在端点效应的问题,采用时间序列分析方法对日长变化序列进行端点延拓,形成一个新序列,然后用新序列建立最小二乘模型,最后再结合最小二乘模型和自回归(Autoregressive,AR)模型对原始日长变化序列进行预报。实验结果表明,在日长变化序列两端增加统计延拓数据,能有效减小最小二乘拟合序列的端点畸变,从而提高日长变化的预报精度,尤其对中长期预报精度提高明显。
基金supported by the West Light Foundation of the Chinese Academy of Sciences
文摘Traditional artificial neural networks （ANN） such as back-propagation neural networks （BPNN） provide good predictions of length-of-day （LOD）. However, the determination of network topology is difficult and time consuming. Therefore, we propose a new type of neural network, extreme learning machine （ELM）, to improve the efficiency of LOD predictions. Earth orientation parameters （EOP） C04 time-series provides daily values from International Earth Rotation and Reference Systems Service （IERS）, which serves as our database. First, the known predictable effects that can be described by functional models-such as the effects of solid earth, ocean tides, or seasonal atmospheric variations--are removed a priori from the C04 time-series. Only the residuals after the subtraction of a priori model from the observed LOD data （i.e., the irregular and quasi-periodic variations） are employed for training and predictions. The predicted LOD is the sum of a prior extrapolation model and the ELM predictions of the residuals. Different input patterns are discussed and compared to optimize the network solution. The prediction results are analyzed and compared with those obtained by other machine learning-based prediction methods, including BPNN, generalization regression neural networks （GRNN）, and adaptive network-based fuzzy inference systems （ANFIS）. It is shown that while achieving similar prediction accuracy, the developed method uses much less training time than other methods. Furthermore, to conduct a direct comparison with the existing prediction tech- niques, the mean-absolute-error （MAE） from the proposed method is compared with that from the EOP prediction comparison campaign （EOP PCC）. The results indicate that the accuracy of the proposed method is comparable with that of the former techniques. The implementation of the proposed method is simple.
