I can calculate cluster analysis by both hysplit and trajstat model. Now I want to put a standard deviation of those cluster in the plot. How can I do that?
Atmospheric variables are not normally distributed, they have long-tailed power law distributions. The effects of statistical multifractal scale invariance can make trajectory analysis problematical. See the attached review as a starting point.The journal website has a correction to a semantic confusion, 780_ftp.pdf