fundamentals of eDNA

My most exciting part of this week’s module on study design was learning about occupancy modeling. I vaguely heard of it in fisheries applications, but not for eDNA until this week’s lecture. Given the vast uncertainty surrounding eDNA, I like the idea that you can use this analysis to account for and quantify some of the detection uncertainty, and it seems like it could be especially useful with fish eDNA work. The discussion led me to think about whether it is more difficult to apply in a metabarcode analysis than a qPCR one, where sequence results might be more affected by all the other DNA present in the sample. It also makes me wonder how good estimates of “actual” presence have to be in order for the detection probability in the model to be accurate. Also, “abundance” in the context of a freshwater system seems more straightforward than in the ocean. Where are the boundaries inside which you count an organism as “present in the system”?