Untangling the relationship between network complexity and ecological stability under climate change is an arduous challenge for theoretical and empirical ecology. Even more so, when considering extreme climatic events. Here, we studied the effects of extreme climatic events (heatwaves) on the complexity of realistic freshwater ecosystems using topological and quantitative trophic network metrics. Next, we linked changes in network complexity with the investigation of four stability components (temporal stability, resistance, resilience, and recovery) of community's functional, compositional, and energy flux stability. We found reduction in topological network complexity to be correlated with reduction of functional and compositional resistance. However, temperature‐driven increase in link‐weighted network complexity increased functional and energy flux recovery and resilience, but at the cost of increased compositional instability. Overall, we propose an overarching approach to elucidate the effects of climate change on multidimensional stability through the lens of network complexity, providing helpful insights for preserving ecosystems stability under climate change.

Significance: Multiple anthropogenic disturbances affect the structure and functioning of communities. Recent evidence highlighted that, after pulse disturbance, the functioning a community performs may be recovered fast due to functional redundancy, whereas community multivariate composition needs a longer time. Yet, the mechanisms that drive the different community recovery times have not been quantified empirically. We use quantitative food-web analysis to assess the influence of species interactions on community recovery. We found species-interactions strength to be the main mechanism driving differences between structural and functional recovery. Additionally, we show that interactions between multiple disturbances appear in the long term only when both species-interaction strength and food-web architecture change significantly.

Microcystis are known for their potential ability to synthesize toxins, mainly microcystins (MCs). In order to evaluate the effects of temperature on chlorophyll a (Chl a), growth, physiological responses and toxin production of a native Microcystis aeruginosa, we exposed the cells to low (23°C) and high (29°C) temperature in addition to a 26°C control treatment. Exponential growth rate was significantly higher at 29°C compared to 23°C and control, reaching 0.43, 0.32 and 0.33day(-)(1) respectively. In addition, there was a delay of the start of exponential growth at 23°C. However, the intracellular concentration of Chl a decreased significantly due to temperature change. A significant increase in intracellular ROS was observed in coincidence with the activation of enzymatic antioxidant catalase (CAT) during the first two days of exposure to 23° and 29°C in comparison to the control experiment, decreasing thereafter to nearly initial values. Five MCs were determined by LC-MS/MS analysis. In the experiments, the highest MC concentration, 205fg [Leu(1)] MC-LR.cell(-1) expressed as MC-LR equivalent was measured in the beginning of the experiment and subsequently declined to 160fg.cell(-1) on day 2 and 70fg.cell(-1) on day 4 in cells exposed to 29°C. The same trend was observed for all other MCs except for the least abundant MC-LR which showed a continuous increase during exposure time. Our results suggest a high ability of M. aeruginosa to perceive ROS and to rapidly initiate antioxidant defenses with a differential response on MC production.

The effect of temperature (26°C, 28°C, 30°C and 35°C) on the growth of native CAAT-3-2005 Microcystis aeruginosa and the production of Chlorophyll-a (Chl-a) and Microcystin-LR (MC-LR) were examined through laboratory studies. Kinetic parameters such as specific growth rate (μ), lag phase duration (LPD) and maximum population density (MPD) were determined by fitting the modified Gompertz equation to the M. aeruginosa strain cell count (cellsmL-1). A 4.8-fold increase in μ values and a 10.8-fold decrease in the LPD values were found for M. aeruginosa growth when the temperature changed from 15°C to 35°C. The activation energy of the specific growth rate (Eμ) and of the adaptation rate (E1/LPD) were significantly correlated (R2=0.86). The cardinal temperatures estimated by the modified Ratkowsky model were minimum temperature=8.58±2.34°C, maximum temperature=45.04±1.35°C and optimum temperature=33.39±0.55°C. Maximum MC-LR production decreased 9.5-fold when the temperature was increased from 26°C to 35°C. The maximum production values were obtained at 26°C and the maximum depletion rate of intracellular MC-LR was observed at 30-35°C. The MC-LR cell quota was higher at 26 and 28°C (83 and 80fgcell-1, respectively) and the MC-LR Chl-a quota was similar at all the different temperatures (0.5-1.5fgng-1). The Gompertz equation and dynamic model were found to be the most appropriate approaches to calculate M. aeruginosa growth and production of MC-LR, respectively. Given that toxin production decreased with increasing temperatures but growth increased, this study demonstrates that growth and toxin production processes are uncoupled in M. aeruginosa. These data and models may be useful to predict M. aeruginosa bloom formation in the environment.

The aim of this study was to evaluate the effects of short-term (hours) exposure to solar UV radiation (UVR, 280–400 nm) on the physiology of Microcystis aeruginosa. Three solar radiation treatments were implemented: (i) PAR (PAR, 400–700 nm), (ii) TUVA (PAR + UVAR, 315–700 nm) and (iii) TUVR (PAR + UVAR + UVBR, 280–700 nm). Differential responses of antioxidant enzymes and the reactive oxygen species (ROS) production to UVR were observed. Antioxidant enzymes were more active at high UVR doses. However, different responses were observed depending on the exposure to UVAR or UVBR and the dose level. No effects were observed on the biomass, ROS production or increased activity of superoxide dismutase (SOD) and catalase (CAT) compared to the control when UVR + PAR doses were lower than 9875 kJ m⁻². For intermediate doses, UVR + PAR doses between 9875 and 10 275 kJ m⁻², oxidative stress increased while resistance was imparted through SOD and CAT in the cells exposed to UVAR. Despite the increased antioxidant activity, biomass decrease and photosynthesis inhibition were observed, but no effects were observed with added exposure to UVBR. At the highest doses (UVR + PAR higher than 10 275 kJ m⁻²), the solar UVR caused decreased photosynthesis and biomass with only activation of CAT by UVBR and SOD and CAT by UVAR. In addition, for such doses, a significant decrease of microcystins (MCs, measured as MC-LR equivalents) was observed as a consequence of UVAR. This study facilitates our understanding of the SOD and CAT protection according to UVAR and UVBR doses and cellular damage and reinforces the importance of UVR as an environmental stressor. In addition, our results support the hypothesized antioxidant function of MCs.