图  1  TIFF格式的MSLA-UV数据（2016年1月1日）

黄色区域为无效值区域，其他不同颜色表示流速差异

Fig.  1  MSLA-UV data in TIFF format（January 1, 2016）

The yellow area is the invalid value area, and other different colors indicate the flow rate difference

图  2  栅格模板数据（2016年1月1日）

黑色为无效值区域（大陆与北极冰雪覆盖地或非涡海域），浅灰色为冷涡，深灰色为暖涡

Fig.  2  Grid template data (January 1, 2016)

Black represents the invalid value area (continent and Arctic snow-covered areas or non-vortex sea areas), light gray represents cold vortex, and dark gray represents warm vortex

图  3  涡旋顶点坐标重构示意图

Fig.  3  Reconstruction of eddy vertex coordinates

图  4  涡旋追踪结果示意图

红色折线是涡心所经过的路径，折线上的黑色圆点为涡心位置

Fig.  4  Schematic diagram of eddy tracking results

The red line is the path through the eddy center, and the black dots on the red line are the position of the eddy center

图  5  二维矢量场时空连续框架示意图（修改自文献[20]）

Fig.  5  Schematic diagram of space-time continuum frame of two-dimensional vector field (modified from reference[20])

图  6  粒子密度随相机变化示意图（视口内播撒的粒子数目相同）

Fig.  6  Schematic diagram of particle density changing with the camera（the number of particles in the viewport are the same）

图  7  单个粒子的影响域（a），两个粒子的影响域叠加（b）和局部的概率密度图（c）

Fig.  7  The influence domain of a single particle (a), the superposition of the influence domain of two particles (b), and the local probability density diagram (c)

图  8  二维流线可视化流程图

a.坐标纹理；b.概率密度纹理；c.颜色表示年龄信息的密度纹理，粒子年龄为0时，颜色为白色，年龄在[0,1]时为黄色，年龄在[2,3]时颜色为红色；d.逐步积分生成流线，$ {\tau }_{0} $为起点，$ \tau $为终点；e.地球表面生成的流线；f.地球表面生成流线的局部放大图

Fig.  8  2D streamline visualization flowchart

a. coordinates texture; b. probability density texture; c. color represents the age information of density texture; when the particle’s age is 0, the color is white; when the age is [0,1], it is yellow; when the age is [2,3], it is red. d. step-by-step generation of streamlines, $ {\tau }_{0} $ and $ \tau $ are the beginning and end of integral time; e. streamlines on the earth’s surface; f. local magnification of streamlines on the earth’s surface

图  9  OW方法（a）、栅格方法（b）和矢量方法（c）的可视化流程图

与流线可视化主要区别在PASS3中的蓝色部分

Fig.  9  Visual flow charts of OW method (a), grid method (b) and vector method (c)

The main difference from streamline visualization is in the blue section of PASS3

图  10  OW方法插值过程示意图

Fig.  10  A schematic diagram of the interpolation process in OW method

图  11  栅格方法插值过程示意图

Fig.  11  A schematic diagram of the interpolation process in grid method

图  12  追踪数据与识别数据的关系示意图

Fig.  12  Schematic diagram of the relationship between tracking data and identifying data

图  13  重构数据的顶点坐标插值示意图

Fig.  13  Interpolation diagram of vertex coordinates of reconstructed data

图  14  传输函数交互实现原理示意图

Fig.  14  Transfer function interaction principle diagram

图  15  3种方法绘制的涡旋可视化结果（2016年1月1日）

第一、二、三行分别为OW方法、栅格方法、矢量方法绘制的结果，第一、二、三列的传输函数分别为W值、涡度和速度。传输函数中W参数和涡度的横坐标值都被放大了10¹⁰倍。涡度值的正号表示反气旋涡，负号表示气旋涡；W参数值大于0的部分被移除，将气旋涡的W参数的绝对值取代了W参数大于0的部分

Fig.  15  Visualization renderings of the three methods (January 1, 2016)

The first, second and third rows are drawn by OW method, grid method and vector method, respectively. The transmission functions of the first, second and third columns are W value, vorticity and velocity, respectively. The x-coordinate values of W parameter and vorticity in the transfer function are magnified 10¹⁰ times. The positive sign of vorticity value means anti-cyclonic and negative represents cyclonic. The part where the value of the W parameter is greater than zero is removed, and the part where the W parameter is greater than zero is replaced by the absolute value of the W parameter of cyclonic

图  16  OW方法（a）、栅格方法（b）、矢量方法（c）绘制的涡旋可视化效果局部放大图

Fig.  16  Local visualization renderings of OW method (a), grid method (b) and vector method (c)

图  17  传输函数交互效果图

Fig.  17  Transfer function interaction diagram

图  18  顶点数目对绘制效率的影响

Fig.  18  The effect of the number of vertices on rendering efficiency

图  19  积分步长对绘制效率的影响

Fig.  19  The influence of integral step on drawing efficiency