Cloud altitude measurements by a 532nm backscatter Lidar and time lapsed digital photography are combined to monitor the cloud velocity profile. The cloud images are recorded in time steps of two seconds by a Nikon D100 digital camera through a 63° solid angle while the Lidar was measuring the cloud altitude. The images are recorded in 8 bits gray scale JPG format in an array of 2240×1488 pixels. To measure the angular displacement of different parts of the cloud, each image is meshed into an array of 44×29 cells, each cell contains 50×50 pixels. The grayscale density cross correlations between similar cells of successive images are computed using a MATLAB code developed by us for this application. The output products are the direction and the amount of displacement of each cell, in pixels. combining the results on cloud displacement with Lidar measurements enable to calculate the velocity vector in each cell. The resolution in velocity is about 1 ms^-1 and 2° in direction. The calculation technique also is tested by simulating the cloud motion by moving the image pixels with a computer generated Gaussian velocity distribution.

Date Published: 31 October 2005
PDF: 9 pages
Proc. SPIE 5979, Remote Sensing of Clouds and the Atmosphere X, 59790C (31 October 2005); doi: 10.1117/12.629436

Published in SPIE Proceedings Vol. 5979:
Remote Sensing of Clouds and the Atmosphere X
Klaus Schäfer; Adolfo T. Comerón; James R. Slusser; Richard H. Picard; Michel R. Carleer; Nicolaos Sifakis, Editor(s)