Motion tracking and cloud height assignment methods for Himawari-8 AMV
Motion tracking and cloud height assignment methods for Himawari-8 AMV
复制标题
Himawari-8 AMV 的运动跟踪和云高分配方法
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
Kazuki Shimoji
中科院分区：
文献类型：
--
作者：
Kazuki Shimoji
日本下一代Himawa-8/9卫星将搭载先进的Himawa Imager(AHI)单元，能够每10分钟产生16个通道的全磁盘图像。对于红外通道，卫星子点的空间分辨率为2公里。这些卫星的观测数据的使用预计将使以高时间、空间和光谱分辨率的数据为基础的先进产品得以产出。日本气象厅气象卫星中心(JMA/MSC)正在为Himawa-8大气运动矢量(AMV)开发新的跟踪和高度估计算法。新算法的设计是为了有效利用AHI的高空间、时间和光谱分辨率。算法的主要变化是应用于云特征跟踪和云高度估计过程。在跟踪方法中，分别准备小目标盒和大目标盒来计算两个关联面。用小目标盒的相关面作先验信息来估计风矢量，用大目标盒的另一个关联面作为辅助信息来确定与小尺度和大尺度大气运动都一致的最佳风矢量。Himawa i-8 AMV的高度估计方法是基于最优估计，以最小化由三个或更多通道的云分配和辐射传输模式参数确定的辐射度观测值与理论值之间的差异。该方法将应用于Himawa-8/9 IR/WV风矢量的高空、中层和低层云。
Japanese next-generation Himawari-8/9 satellites will carry Advanced Himawari Imager (AHI) units capable of producing full-disk images every 10 minutes with 16 channels. The spatial resolution at the sub satellite point (SSP) is 2 km for IR channels. The use of observation data from these satellites is expected to enable the output of advanced products based on data with high temporal, spatial and spectral resolutions. The Meteorological Satellite Centre of the Japan Meteorological Agency (JMA/MSC) is developing new tracking and height estimation algorithm for Himawari-8 Atmospheric Motion Vector (AMV). The new algorithm is designed for effective utilization of high spatial temporal and spectral resolutions of AHI. Major changes of the algorithm are applied to cloud feature tracking and cloud height estimation process. In the tracking method, small and large target boxes are prepared respectively for computing two correlation surfaces. Correlation surface from small target box is used as prior information for estimating wind vector, and another correlation surface derived from large target box is used as auxiliary information for determining optimal wind vector which is consistent with both of small and large scale atmospheric motion. Approach to height estimation method for Himawari-8 AMV is based on optimal estimation to minimize the difference between observed radiance values and the theoretical ones determined from cloud assignment and radiative transfer model parameters using three or more channels. The method will be applied to upper-, medium and low-level clouds for Himawari-8/9 IR/WV wind vectors.