Entomologists and ecologists

Community and Forum → Other questions. Insects topics → Entomologists and ecologists

Elena 2012, 09.09.2013 13:53

Hello dear forumchane!!!I wanted to ask for some advice.I'm writing a dissertation on ecology.A community of phytophagous insects
(cruciferous fleas, cabbage moth, cabbage scoop, cabbage and turnip whitefly, cabbage aphid)
was studied in the agrocenosis of cabbage on THREE VARIETIES (white cabbage, red cabbage and cauliflower), in addition to entomophagous aphids (sirphids, aphidiids and gallica), all dynamics over 5 years.
I made graphs on the dynamics of the number of these insects. I wanted to calculate the indices of Jaccard, Macintosh, but they do not fit, because aphids were taken into account in points and percentages, it is not possible to translate it into instances :-).
What else can you calculate?Maybe someone will share some ideas :-)

Comments

Likes: 1

11.09.2013 19:09, PVOzerski

God be with them, with the goals of all work... Well, what is the purpose of calculating these coefficients? After all, they will probably go further into some kind of comparative analysis or cluster analysis?

Logically (although this is "offhand", without experimental verification), the Macintosh index should be able to work with points-if certain conditions are met. First, the same number of points should correspond to approximately the same number of individuals in all the compared species (and if the condition is not met, it is pointless to count diversity at all). Second, the scale should be uniform, meaning that an abundance of 1 point differs from an abundance of 2 points about as much as an abundance of 2 points differs from an abundance of 3 points). If this condition is not met, then either the scale needs to be linearized somehow, or the variety will not be calculated correctly (although, say, a comparative analysis based on some nonparametric may still be possible).

Apparently, the same is true for similarity indices. But in my opinion, this problem does not apply to Jaccard at all: you probably know the number of species not in points and not in percentages.