China is the largest dam-building country in the world, with the total estimated storage capacity of reservoirs being triple that of natural lakes. Reservoirs are highly dynamic environments and changes in the hydrological characteristics and shifts in ecosystems state may threaten biodiversity and cause large losses of sustainable ecosystem goods and services. Phytoplankton assemblages are key components in determining ecosystem stability, and they are among the classic examples of populations undergoing switches between alternative stable states in shallow lakes. Shifts in phytoplankton community are often thought to be a result of changes in nutrient load and light regime. However, strong shifts in phytoplankton communities may also be associated with hydrological changes. One of the major factors is water level fluctuations (WLF), which are controlled by both climate changes and human activities. Previous studies have largely focused on unraveling the mechanisms behind regime shifts in shallow lakes or exploring WLF and other environmental variables (such as nutrients and temperature) as important drivers of phytoplankton community structure in freshwater ecosystems. However, the extent to which WLF is a predictor of cyanobacteria dynamics in subtropical reservoirs is still unclear.
The Aquatic Ecohealth (Dr. Jun YANG) Group in IUE, CAS examined effects of water level fluctuations (WLF) on phytoplankton communities with emphasis on cyanobacteria dynamics in four subtropical drinking water reservoirs in southeast China. Compared with temperate reservoirs, these subtropical reservoirs were characterized by different succession patterns of phytoplankton composition. One pattern was dominated by cyanobacteria with high turbidity and high biomass at low negative WLF, and the other dominated by non-cyanobacterial taxa associated with clear water and low biomass at high positive WLF. WLF can influence cyanobacteria biomass directly or indirectly by influencing the light availability, nutrient loads, and other limnological parameters, therefore making it a significant though not necessarily strong predictor of cyanobacteria biomass in these reservoirs. Positive WLF (i.e. an increase in water level) can reduce the cyanobacteria biomass and dominance in complex ways. They also found that to control the risk of cyanobacterial dominance or blooms, WLF should be targeted to be above +2 m/month; that is an increase in water level of 2 m or more. Given that cyanobacterial blooms are a worldwide environmental problem, coupled with the fact that WLF is likely to be more frequent under increased artificial manipulation and more extreme climatic variability, water level management can be widely used in small and medium-sized reservoirs to prevent the toxic cyanobacterial blooms and to protect the ecosystem integrity or functions.
This study has been published in Science of the Total Environment (557/558: 445–452, 2016), entitled “Decline in water level boosts cyanobacteria dominance in subtropical reservoirs.” The work was financially supported by the National Basic Research Program of China (2012CB956103), the National Natural Science Foundation of China (31370471 and 31500372), and the Natural Science Foundation for Distinguished Young Scholars of Fujian Province (2012J06009).
Article link: http://www.sciencedirect.com/science/article/pii/S0048969716305186

Conceptual diagram of the effect of climate change and human activity on water level fluctuations (WLF) and phytoplankton biomass (mean biomass ± SE). Light blue shading indicates uncertainty of WLF. Green pie charts represent the percentage of cyanobacteria to total phytoplankton biomass.

ABSTRACT
Globally aquatic ecosystems are likely to become more vulnerable to extreme water fluctuation rates due to the combined effects of climate change and human activity. However, relatively little is known about the importance of water level fluctuations (WLF) as a predictor of phytoplankton community shifts in subtropical reservoirs. In this study, we used one year of data (2010–2011) from four subtropical reservoirs of southeast China to quantify the effects of WLF and other environmental variables on phytoplankton and cyanobacteria dynamics. The reservoirs showed an apparent switch between a turbid state dominated by cyanobacteria and a clear state dominated by other non-cyanobacterial taxa (e.g., diatoms, green algae). Cyanobacterial dominance decreased, or increased, following marked changes in water level. Multiple regression analysis demonstrated that pH, euphotic depth, WLF, and total phosphorus provided the best model and explained 30.8% of the variance in cyanobacteria biomass. Path analysis showed that positive WLF (i.e. an increase in water level) can reduce the cyanobacteria biomass either directly by a dilution effect or indirectly by modifying the limnological conditions of the reservoirs in complex pathways. To control the risk of cyanobacterial dominance or blooms, WLF should be targeted to be above +2 m/month; that is an increase in water level of 2 m or more. Given that WLF is likely to be of more frequent occurrence under future predicted conditions of climate variability and human activity, water level management can be widely used in small and medium-sized reservoirs to prevent the toxic cyanobacterial blooms and to protect the ecosystem integrity or functions.
Keywords: water level fluctuations, phytoplankton, alternative stable states, climate change, human activity, resilience Corresponding author: Tel. / Fax: +86 5926190775. E-mail address: jyang@iue.ac.cn (Jun Yang)