We performed a transplant experiment in eutrophic Sau reservoir to assess the factors that control bacterial abundance, activity, growth rate, and community composition. Samples from the lacustrine and the riverine ends of the reservoir were incubated in dialysis bags placed in situ and transplanted to the other side of the reservoir and also incubated after 1 m filtration to measure predator effects. The bags were sampled at 12-h intervals to estimate bacterial abundance, whole community activity, activity structure (by flow cytometry), and phylogenetic composition (by in situ hybridization with group-specific phylogenetic probes). Bacterial production was always regulated by nutrient supply, but abundance and activity were differently regulated at both sites. The riverine bacteria were limited by predator activity, whereas the lacustrine were regulated by a combination of predation and nutrient supply. Therefore, even in the same environment, different modes of control can act simultaneously. Bacterial activity structure was also regulated in the same way. Abundance of highDNA bacteria and cells hybridizing with the universal EUB338 probe were well correlated. In the lacustrine sample, bacterial community structure did not change significantly, whereas in the riverine sample, ␣-and ␥-Proteobacteria reduced their growth when transplanted, whereas ␤-Proteobacteria were stimulated by the presence of predators. Members of the Cytophaga/Flavobacterium phylum grew only when incubated in situ in the absence of predators. This different behavior in the different bacterial groups resulted in strong changes in bacterial assemblage composition, evident already after 24 h. The experiment demonstrates that, together with the effect of predators, nutrient supply affects bacterial community properties and that a complex regulation involving both types of control can occur in a single heterogeneous planktonic system. Bacteria are relevant members of the limnetic planktonic food web, both in terms of biomass and production share (e.g., Cole and Caraco 1993;del Giorgio and Gasol 1995) and while this is very evident in oligotrophic plankton sys-1 Corresponding author

No abstract

No abstract

A B S T R A C TWe studied changes in the epilimnetic bacterial community composition (BCC), bacterial biomass and production, and protistan succession and bacterivory along the longitudinal axis of the canyonshaped, highly eutrophic Sau Reservoir (NE Spain) during two sampling campaigns, in April and July 1997. Longitudinal changes in BCC from the river inflow to the dam area of the reservoir were detected by using oligonucleotide probes targeted to the kingdom Bacteria, to the alpha, beta, and gamma subclasses (ALFA, BETA, and GAMA) of the class Proteobacteria, and to the Cytophaga/ Flavobacterium (CF) cluster. In general, the inflow of the organically loaded Ter river, with highly abundant allochthonous bacterial populations, induced a clearly distinguishable longitudinal succession of the structure of the microbial food web. The most dynamic changes in microbial parameters occurred at the plunge point, the mixing area of river water and the reservoir epilimnion. Changes within members of BETA and CF were the most important in determining changes in BCC, bacterial abundance and biomass. Much less relevant changes occurred within the less abundant ALFA and GAMA bacteria. From the plunge point downstream, we described a significant shift in BCC in the form of decreased proportions of BETA and CF. This shift spatially coincided with the highest values of heterotrophic nanoflagellate bacterivory (roughly doubled the bacterial production). CF numerically dominated throughout the reservoir without any marked longitudinal changes in their mean cell volume. In contrast, very large cells affiliated to BETA clearly dominated in the allochthonous bacterial biomass brought by the river. BETA showed a marked downstream trend of decreasing mean cell volume. We conclude that the observed BCC shift and the longitudinal shift in food web structure (bacteria-heterotrophic nanoflagellates-ciliates) resulted from highly complex interactions brought about by several major factors: varying hydrology, the high localized allochthonous input of organic matter brought by the river, downstream changing substrate availability, and selective protistan bacterivory.

Low dissolved oxygen concentration in bottom layers of lakes and reservoirs usually indicates low water quality. In lakes, empirical models predicting anoxia are almost entirely based on the decay of plankton biomass, while in reservoirs recent findings suggest a prominent role of streamflow and load of organic carbon. This suggests a potential link between water quality in reservoirs and climate processes affecting streamflow. Here we support this hypothesis presenting evidence that both interannual climate variability and recent climate change, mainly consisting in a significant increase in potential evapotranspiration in the upstream basin, affected the oxygen content in a Mediterranean reservoir (Sau Reservoir, Spain). Using a 44-year monthly record, we found strong and consistent signatures of El Niñ o Southern Oscillation in the inflow and reservoir oxygen content. Spectral and wavelet techniques showed that the El Niñ o, streamflow, and reservoir oxygen content series oscillated in common periods, which coincided with the main El Niñ o variability modes. An empirical model explaining the annual oxygen content in the reservoir suggested that a decreasing streamflow trend reduced the oxygen content of the reservoir by about 20%, counteracting remediation measures implemented at the basin upstream the reservoir. Our results provide the first quantitative evidence of climate change effects on reservoir water quality using long-term instrumental data, and indicate that streamflow should be considered as a key variable in assessing climate change impact on reservoir water quality. These results are especially relevant in regions of the world where reservoirs are abundant and most climate models predict a decrease in runoff during the next decades. Both the expected trends and the sensitivity of reservoir water quality to global interannual climate variability should be considered for a correct management of water resources in the present and to design adaptation policies in the future.