adonis2 for a study with unbalanced presence/absence data

[JELLY1]

Hi, I am evaluating a dataset of 6 different field sites (hereby called "Block" whereby we collected data at multiple plots within each site (though the number of plots was not even - I joined the project post-data collection). Due to the nature of how data was collected I am just using presence/absence of each species within each sampling effort. Each plot was sampled multiple times (again not an even number always due to weather, access, etc). So I have a very unbalanced design. Despite this challenge, I am trying to determine if the species composition of the sites differs from one another. In addition, I am trying to determine if these differences are due to different environmental variables: human pop density, canopy cover, percent water cover.

I have been using method = "bray" with binary = TRUE to account for our presence/absence data collection with the following code based on what I could find here in the forum and literature:

h2 <- with(birds.com, how(nperm = 999,
                          plots = Plots(strata = bird.ev$Block, type = "free"),
                          within = Within(type = "none")))

adonis2(formula = birds.com ~ canopy.inside , data = bird.ev, by='margin', method = "bray",binary=TRUE,permutations = h2) ##BC

but I get the error:

Error in check(sn, control = control, quietly = quietly) : 
  Design must be balanced if permuting 'strata'.

Now I can run the following but essentially setting Blocks as strata argument which as far as I understand then permutes within the blocks.

h1 <- with(birds.com, how(nperm = 999, blocks = bird.ev$Block))
adonis2(formula = birds.com ~ canopy.inside , data = bird.ev, by='margin', method = "bray",binary=TRUE,permutations = h1) ##BC

So my questions are:

1. What is the best way to evaluate whether the blocks differ from one another with regards to species composition?
2. Do particular environmental covariates drive differences in species composition between the Blocks?

Thank you very much for you help
Amanda