Are cyanobacteria total, specific and trait abundance regulated by the same environmental variables?

Abstract

In this study we analyzed if cyanobacteria total, specific and trait abundance are regulated by the same environmental variables in a Neotropical urban lake that recurrently suffers harmful cyanobacteria blooms. To assess the predictor variables for cyanobacteria total and species density we performed a multiple regression (GLM) and a redundancy analysis (RDA), respectively. Temperature and oxygen were the main predictor variables for both total and species abundance. Conductivity was an exclusive predictor for cyanobacteria total density (GLM) and light availability (Zd:Zeu) for species abundance (RDA). Nutrients were unnoticeable predictor variables for both. Cyanobacteria blooms showed high recurrence (8 blooms in 12 months) and occurred within 17-28 °C. Blooms were mostly dominated by one species, and less frequently co-dominated by two species. These blooms were more recurrently dominated by dispersive non-fixing filamentous species (mainly Raphidiopsis curvata) linked to lower light availability. Less frequently, blooms were dominated by filamentous nitrogen fixers which develop scum blooms (mainly Anabaenopsis arnoldii) related to better light availability and lower dissolved oxygen concentration. The nitrogen fixing species showed high heterocyte density, suggesting nitrogen fixing behavior and probably giving this an advantage when inorganic nitrogen was low. Our results indicate that in absence of nutrients limitation, cyanobacteria total and species abundance can be regulated by different environmental variables. These results also show that species phylogenetically related (R. curvata and A. arnoldii) can respond differently to the prevailing environmental variables; highlighting the importance of considering cyanobacteria to a specific level when assessing their possible control factors.