The general regression neural network(GRNN) model was proposed to model and predict the length of day(LOD) change, which has very complicated time-varying characteristics. Meanwhile, considering that the axial atmospheric angular momentum(AAM) function is tightly correlated with the LOD changes, it was introduced into the GRNN prediction model to further improve the accuracy of prediction. Experiments with the observational data of LOD changes show that the prediction accuracy of the GRNN model is 6.1% higher than that of BP network, and after introducing AAM function, the improvement of prediction accuracy further increases to 14.7%. The results show that the GRNN with AAM function is an effective prediction method for LOD changes.
Empirical functional models for the maximum and minimum detectable deformation gradient of PALSAR interferometry were established based on coherence and discrete look numbers. Then, a least square regression method was used to fit the model coefficients and thus obtain the generalized functional models for both coherence and look numbers. The experimental results with ALOS PALSAR data of Wenchuan earthquake of China show that the new model works well for judging whether the deformation gradient can be detected by the D-InSAR technology or not. The results can help researchers to choose PALSAR data and to configure processing parameters, and also benefit the interpretation of the measured surface deformation.
The Malkmus band model has been widely used in remote sensing and climate studies. However, its accuracy is not high. To solve this problem, a modified Malkmus band model was proposed by introducing a correction item. The HITRAN (High-resolution TRANsmission) 2008 database and the atmospheric models provided by the Air Force Geophysics Laboratory (AFGL) were used to calculate the molecular transmittances. By fitting the calculated transmittances to those by MODTRAN (MODerate resolution atmospheric TRANsmission) package with the least-squares method, the fitting coefficients of the correction item were obtained under different atmosphere models. The experimental results show that the root mean square errors (RMSE) of the modified model are significantly less than that of the traditional Malkmus band model by 1-2 orders of magnitude. In addition, the modified method is suitable for different atmospheric models and molecules.
