The TOMS zonal average total ozone data in the Northern Hemisphere are decomposed with the empirical orthogonal function (EOF) method. According to the features of the spatial characteristic vectors, the characteristi...The TOMS zonal average total ozone data in the Northern Hemisphere are decomposed with the empirical orthogonal function (EOF) method. According to the features of the spatial characteristic vectors, the characteristic vectors that have been obtained with EOF method can be used as the ordered orthogonal radixes to unfold the phase space. After the corresponding time functions are embedded in phase space, the traces of the state vectors of the regional ozonosphere dynamical system are constructed, and can be used to describe the attractor integral information of the asymptotic state of the regional ozonosphere system and the dynamical features of the regional ozonosphere system, and then the embedded saturation dimension of the regional ozonosphere system attractor is successfully obtained. Based on these works mentioned above, by using the time function series we solve a problem contrary to the numerical solution and retrieve the control parameters of the state equations in which quadratic nonlinear terms are included, and then the dynamical models that can objectively reflect the temporal variation of the regional ozonosphere system are finally established.展开更多
文摘The TOMS zonal average total ozone data in the Northern Hemisphere are decomposed with the empirical orthogonal function (EOF) method. According to the features of the spatial characteristic vectors, the characteristic vectors that have been obtained with EOF method can be used as the ordered orthogonal radixes to unfold the phase space. After the corresponding time functions are embedded in phase space, the traces of the state vectors of the regional ozonosphere dynamical system are constructed, and can be used to describe the attractor integral information of the asymptotic state of the regional ozonosphere system and the dynamical features of the regional ozonosphere system, and then the embedded saturation dimension of the regional ozonosphere system attractor is successfully obtained. Based on these works mentioned above, by using the time function series we solve a problem contrary to the numerical solution and retrieve the control parameters of the state equations in which quadratic nonlinear terms are included, and then the dynamical models that can objectively reflect the temporal variation of the regional ozonosphere system are finally established.
