太阳深积分磁场观测中异常结构的改正

Correcting false structures in solar deep-integration magnetogram observations

2012年,中国科学院国家天文台怀柔太阳观测基地建成了基于位移叠加技术的深积分实时太阳磁场观测系统,但随着观测的开展,发现在计算的磁图中有时会存在整体性的异常颗粒结构,这些结构显然不是太阳磁场结构.本文首先对其成因进行了仔细分析,认为这是由于在积分计算磁场数据的时候,采用了第1,2帧作为左右旋偏振图像的基准帧引起的,这两帧图像由于大气抖动或者其他原因,彼此会有一定的相对位移,而这种位移在左旋和右旋2个偏振态的数据相减的时候,就会产生颗粒状的异常结构.为此,本文提出了使用左右旋平均视场来代替单帧图像作为基准图像的方法,并给出了详细的实现方法.通过多角度的实验和数据分析,图像的视觉效果和数据统计结果都表明本文的改正方法简单且有效.

The real-time correlation solar observation system with a deep integration technique has been operating since 2012 at the Huairou Solar Observation Station at the National Astronomical Observatories of China. However, some magnetograms show false granular structures. After analyzing these images, we believe that the false structures are generated in the last step of the magnetic field measurement： the subtraction of the right circular polarized accumulation image from the left one. Since the left or right image is aligned with the first or second frames by a shift and add technique, separately; the intensity difference between the fields of view （FOV） of the first and second frames cause the false structures. We propose a correction method. Two average FOVs of circular polarized accumulation images are selected as the reference images to replace the first and second frames. Since the average FOVs are determined by the displacement which was calculated and recorded during the alignment procedure, this method can be easily implemented in routine observation procedures. The experimental results prove that our method is successful and effective.