What would be the best method for analysing trends in species abundance in a single area in Latin America over a 15 year period? Could TRIM be used for this purpose?
I don't have data about several point-counts in the area over the years. All I have is total abundace per species per year in the area.