search

tokami / TropFishR

Tropical Fisheries Analysis with R
24 stars 19 forks source link

One cohort analysis #30

Closed freddycus closed 3 years ago

freddycus commented 5 years ago

I am trying to apply the ELEFAN routine to jellyfish, their lifespan oscilates typically from 1 to 4 months making it hard to differenciate between cohorts as some of them might be spaced just for 15 days instead of a whole year. However, for one of the species I am completely sure that for the whole lifespan I only have 1 cohort but when I throw the analysis I receive a ncohort of 4, being it extremely high given the loleny recruitment peak registered for the data. Is it possible to fix the ncohort before running ELEFAN so I can get a K value more consistent with the input data?

marchtaylor commented 5 years ago

Hello, Although I haven't used it for a long time, the agemax argument should fix the number of cohorts that are fit by ELEFAN (used internally by lfqFitCurves). That said, I don't think we ever envisioned it to be used in an absolute way - more as a way to standardize the value between fittings using different settings. Looking at the code for that function, it may round up to 2 cohorts when defining agemax = 1. Perhaps, agemax = 0 would work in your case where they live less than a year. If you can give this a try with your data and get back to us, we would be most appreciative.

freddycus commented 5 years ago

Hello, Dr. Taylor. Fixing the agemax = 1 or a fraction of year has been useful for the calculation with my data. In case someone find him/herself in the same situation I wouldn't recommend to use agemax=0 because it makes ELEFAN uncapable of providing with a suitable K value given that all of them obtain the same score. Thanks a lot for your help