I sampled algae from three wetland sites on 13 sampling events. At each wetland site on each sampling event, I collected one sample of algae from the same location at the centre of the wetland and 4 random samples of algae around the littoral zone (wetland margin). So for each event, there is 1 centre sample vs. 4 random littoral samples in each wetland site.
For each wetland site, I ran a non-metric multidimensional scaling (NMDS; PC-ORD) of taxon abundance × [site by event] matrix to visualise differences in community composition among zone types (centre vs. littoral) throughout the study.
I would like to explore the amount of temporal variability between the centre and the littoral zone algal communities throughout the study for each wetland site. I would like to calculate the total Euclidean distance between successive event samples for the centre and for the littoral zone in ordination space using the two NMDS axes in the final solution. This method would be fine for the central zone, since I sampled the same centre location for each event. But for the 4 randomly selected sites in the littoral zone on each event, how do I calculate the Euclidean distance when the 4 littoral sites are not the same for each sampling event. Would it be acceptable to calculate 4 Euclidean distances (even though the sites are random) and calculate the mean Euclidean distances with SE for the 4 littoral sites? Would calculating 16 possibe pairs of Eucli. distances suffice?
Hi Luisa
1. I don't see the need to use euclidean distances from an ordination to look at compositional variation. NMDS is based on some measure of dissimilarity (Bray-Curtis is a popular option). Using the original dissimilarity measures seems like a more direct way of looking at this.
2. Differences between the random samples will be due to both space and time. You could explore this by comparing differences between random samples for a particular time interval (space only) with differences between random samples across time intervals (space and time).
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix