Hi all,

I am currently using a combination of R libraries (ape, vegan, plyr, dplye) to derive a community composition analysis but am failing. First, this data occurs over 5 years so my data frame is arranged by Year/Month/Site identifiers for rows and species by columns. Second, there are a lot of zeros in this data frame that I tried to take care of by applying a log+1 but I don't think it made much difference (at least I don't see where it did). Really what I need to know is 1) should I separate the whole data frame by years and run an analysis for each one, then do PERMANOVA? or 2) is there a way to use an additional column (I would add a year column) to "group" them? I am plotting successfully but having trouble interpreting and believing the analysis is really taking time into account. Any help/advice would be greatly appreciated. I would be happy to provide code/images if that helps.