Hello,
My hypothesis is that the diet composition of a small fish varies between 9 stands of macrophytes, being 3 of low diversity (shannon), 3 of intermediate diversity and 3 of high diversity of macrophytes. Each macrophyte stand was sampled independently along 20km of a river. To test the composition of the diet, I ran a PERMANOVA. So, considering that other unmeasured factors may influence diet composition within the group and between groups (e.g. site location, geographic distance, plant identity), how can I test the differences of diet composition between the groups?
Thank you very much! I hope I wrote it clear enough.
If you have the full environmental data, check firat how they differ between your sites. If the environment is considered heterogeneous and could be affecting the fishes' diet you can add these data as enviromental covariates when running permanova, you will get reports for those separately. Anither option would be to use the envir data to produce a spatial pca, and use the pca values as your covariates. It all depends on number of variables, samples and how homogeneous the habitat is. Let me know if this was helpful
To know if there are differences in the composition of the diet of fish between macrophyte groups and I suppose that there is an analysis of stomach content by measuring any of these variables:% FO,% G,% N,% V and% IIR , in addition to measuring certain abiotic or biotic factors such as time of year, sites etc. I recommend you to do a Two-way PERMANOVA Test, you can do it in Past 3.25. Defining the factors (site, plants, etc.), for a quantitative variable measured.
Another way that I recommend doing it is to establish which variable of stomach content you are going to use as the previous ones that explain, for me the most important is % IIR or abundance or weight of the consumed entities, you can make a Bray-Curtis Cluster. Then draw a cut line to the percent similarity that you define and see how many groups were formed in your cluster. Then he carries out an ANOSIM One-way Test (PERMANOVA), to find out if the groups formed are statistically different.
Attached the manual of Past 3.25 so you can analyze the statistical tests.
You are just assessing composition of diet in relation to macrophyte diversity, so you don't need to worry about other factors imo
Thank you all for your answers. But now I'm in doubt because of the different answers, which all seems plausible, because my initial thoughts were concerning about running only a one way PERMANOVA, considering the 9 stands of plants grouped into the factor diversity that I definied, so 3 groups (low, intermediate and high) with 3 stands for each, or to consider another variable, such as the site or plant identity, since I grouped different stands of macrophytes based only on their diversity. My concern is, because these stands are from different locations in the river, can I grouped them based only in the diversity that I calculated?
- Badges
- Science topic
- Similar topics
- Biological Science
- Zoology
- Ichthyology
- Vertebrates
- Fish