Inter- and intra-annual variability in the phytoplankton community of a high mountain lake: the influence of external (atmospheric) and internal (recycled) sources of phosphorus

1. The inter‐ and intra‐annual changes in the biomass, elemental (carbon (C), nitrogen (N) and phosphorus (P)) and taxonomical composition of the phytoplankton in a high mountain lake in Spain were studied during 3 years with different physical (fluctuating hydrological regime) and chemical conditions. The importance of internal and external sources of P to the phytoplankton was estimated as the amount of P supplied via zooplankton recycling (internal) or through ice‐melting and atmospheric deposition (external). 2. Inter‐annual differences in phytoplankton biomass were associated with temperature and total dissolved phosphorus. In 1995, phytoplankton biomass was positively correlated with total dissolved phosphorus. In contrast, the negative relationship between zooplankton and seston biomass (direct predatory effects) and the positive relationship between zooplankton P excretion and phytoplankton biomass in 1997 (indirect P‐recycling effects), reinforces the primary role of zooplankton in regulating the total biomass of phytoplankton but, at the same time, encouraging its growth via P‐recycling. 3. Year‐to‐year variations in seston C : P and N : P ratios exceeded intra‐annual variations. The C : P and N : P ratios were high in 1995, indicating strong P limitation. In contrast, in 1996 and 1997, these ratios were low during ice‐out (C : P < 100 and N : P < 10) and increased markedly as the season progressed. Atmospheric P load to the lake was responsible for the decline in C : P and N : P ratios. 4. Intra‐annual variations in zooplankton stoichiometry were more pronounced than the overall differences between 1995 and 1996. Thus, the zooplankton N : P ratio ranged from 6.9 to 40.1 (mean 21.4) in 1995, and from 10.4 to 42.2 (mean 24.9) in 1996. The zooplankton N : P ratio tended to be low after ice‐out, when the zooplankton community was dominated by copepod nauplii, and high towards mid‐ and late‐season, when these were replaced by copepodites and adults. 5. In 1995, the minimum demands for P of phytoplankton were satisfied by ice‐melting, atmospheric loading and zooplankton recycling over 100%. In order of importance, atmospheric inputs (> 1000%), zooplankton recycling (9–542%), and ice‐melting processes (0.37–5.16%) satisfied the minimum demand for P of phytoplankton during 1996 and 1997. Although the effect of external forces was rather sporadic and unpredictable in comparison with biologically driven recycle processes, both may affect phytoplankton structure and elemental composition. 6. We identified three conceptual models representing the seasonal phosphorus flux among the major compartments of the pelagic zone. While ice‐melting processes dominated the nutrient flow at the thaw, biologically driven processes such as zooplankton recycling became relevant as the season and zooplankton ontogeny progressed. The stochastic nature of P inputs associated with atmospheric events can promote rapid transitional changes between a community limited by internal recycling and one regulated by external load. 7. The elemental composition of the zooplankton explains changes in phytoplankton taxonomic and elemental composition. The elemental negative balance (seston N : P < zooplankton N : P, low N : P recycled) during the thaw, would promote a community dominated by species with a high demand for P (Cryptophyceae). The shift to an elemental positive balance (seston N : P > zooplankton N : P, high N : P recycled) in mid‐season would skew the N : P ratio of the recycled nutrients, favouring dominance by chrysophytes. The return to negative balance, as a consequence of the ontogenetic increase in zooplankton N : P ratio and the external P inputs towards the end of the ice‐free season, could alleviate the limitation of P and account for the appearance of other phytoplankton classes (Chlorophyceae or Dinophyceae).