I don’t think this was covered in any of the videos.
I am thinking of including an interaction between sponge habitats and fishing in my model. For this interaction we can be quite confident that if fishing is increasing sponge biomass decreases, but because of the life history of sponges, confidence that sponge biomass will increase if fishing decreases is much lower. It looks like it is possible to change the probabilities in the RecipProbs2 and RecipProbs3 tab in the excel template to cover this situation, and the model behaves as expected but I wanted to double check?
Also, are there any suggestions on how to clearly represent two probabilities for the same interaction within an interaction diagram?
I dont quite get your question. But in my understandings, the interaction diagram is intent to represent interactions (either competitive or top-down) between nodes and help you fill in the interaction grid tab in the BBN excel platform from which you will assign probabilities. As such, it would be unnecessary to represent two probabilities for the same interaction in an interaction diagram if this is what you meant?
Thanks for the comments – yes, you are right, you can look at ‘unbalanced’ interactions by altering the recipricol probability tabs. RecipProb2 would be the probability of a ‘column’ node decreasing, assuming the ‘row’ node is increasing, And RecipProb3 is the probability of a column node increasing given the row node is decreasing. By adjusting this, you can account for differences in things like growth rates of species (which will work as suggested for sponges). I did this in the first of the papers (on rocky shores) as the different seaweeds had considerably different growth rates.
You are right though – I don’t cover it in the videos… Sorry