I am performing a db-RDA in R using the package Vegan and the function capscale. I have taken into account 24 environmental factors but the program only considerate 7 of this factors. I have 54 samples but some of them sharing the same environmental value. Do you know why is the problem and don´t considerate all the environmental factors?
thanks in advance
Not exactly my expertise, but as I understand, the distance based RDA selects the most appropriate parameters to use for RDA. This explains in more detail. 24 variables is a lot to analyze and expect meaningful relationships to emerge.
https://sites.google.com/site/mb3gustame/constrained-analyses/rda/dbrda
Thank you so much and another question if you can answer to me. It is more correct to include variables (for example, to associate the variables of ammonium, nitrite and nitrate to total nitrogen?)
@ Sergio, I believe the program you mentioned in your first question has Principal Component Analysis (PCA) incorporated. This enables it to select the Principal Components from the 24 Environmental factors.
The function capscale should include all your variables unless you do a prior selection, which is the recommended. Could you share your code?
Thank you very much, I found the solution. The problem is that it put more environmental variables than the response variables that I had. I have chosen to make a previous selection of the variables as you comment :)
- Badges
- Science topic