Toxic cyanobacterial blooms have become a severe global hazard to human and environmental health. Most studies have focused on the relationships between cyanobacterial composition and cyanotoxins production. Yet, little is known about the environmental conditions influencing the hazard of cyanotoxins. Here, we analysed a unique 22 sites dataset comprising monthly observations of water quality, cyanobacterial genera, zooplankton assemblages, and microcystins (MCs) quota and concentrations in a large-shallow lake. Missing values of MCs were imputed using a non-negative latent factor (NLF) analysis, and the results achieved a promising accuracy. Furthermore, we used the Bayesian additive regression tree (BART) to quantify how Microcystis bloom toxicity responds to relevant physicochemical characteristics and zooplankton assemblages. As expected, the BART model achieved better performance in Microcystis biomass and MCs concentration predictions than some comparative models, including random forest and multiple linear regression. The importance analysis via BART illustrated that the shade index was overall the best predictor of MCs concentrations, implying the predominant effects of light limitations on the MCs content of Microcystis. Variables of greatest significance to the toxicity of Microcystis also included pH and dissolved inorganic nitrogen. However, total phosphorus was found to be a strong predictor of the biomass of total Microcystis and toxic M. aeruginosa. Together with the partial dependence plot, results revealed the positive correlations between protozoa and Microcystis biomass. In contrast, copepods biomass may regulate the MC quota and concentrations. Overall, our observations arouse universal demands for machine-learning strategies to represent nonlinear relationships between harmful algal blooms and environmental covariates.

Background A huge reservoir was formed by the Three Gorges Dam in China, which also formed a riparian zone along the bank of the reservoir. In the period of low water-level, the riparian zone in tributary bays of the Three Gorges Reservoir (TGR) was always unordered cultivated, owing to its gentle slope and high soil fertility. This land-use practice creates high potential of generating greenhouse gas (GHG) emissions with periodic water level fluctuation. Methods To evaluate potential GHG emissions from the soil-air interface, the static opaque chamber method was adopted to evaluate the effect of elevations (180 m, 175 m, 170 m and 165 m) and land use types (dry lands, paddy fields and grass fields) from April to September in 2015 and 2016. Results The results showed that carbon dioxide (CO₂) was the main contributor of GHG emission in riparian zone most likely because of high organic carbon from residues. Furthermore, high soil water content in paddy fields resulted in significantly higher methane (CH₄) flux than that in dry lands and grass fields. Compared to grass fields, anthropogenic activities in croplands were attributed with a decrease of soil total carbon and GHG emissions. However, inundation duration of different elevations was found to have no significant effect on CH₄ and CO₂ emissions in the riparian zone, and the mean nitrous oxide (N₂O) flux from dry lands at an elevation of 165 m was significantly higher than that of other elevations likely because of tillage and manure application. The high N₂O fluxes produced from tillage and fertilizer suggested that, in order to potentially mitigate GHG emissions from the riparian zone, more attention must be paid to the farming practices in dry lands at low elevations (below 165 m) in the riparian zone. Understanding factors that contribute to GHG emissions will help guide ecological restoration of riparian zones in the TGR.

Hydrological regime is widely recognized as the primary driver shaping riparian vegetation, yet its mechanistic links with taxonomic, functional, and ecosystem diversity under dam-regulated conditions remain insufficiently quantified. In this study, we quantified flow regime characteristics in the Three Gorges Reservoir (TGR). We identified five statistically distinct flow regime types using hierarchical clustering based on magnitude, duration, frequency, average flooding intensity, and rate of change. Significant differences among the five flow regime types were observed using one-way ANOVA with Tukey HSD post hoc tests (p < 0.05), particularly in magnitude, duration, flooding intensity, and rate of change, while flooding frequency showed no significant variation. Species richness was negatively associated with flooding duration but positively associated with hydrological conditions of milder flow regime types, especially during early and late growing seasons (April and September). Functional diversity increased along the flow regime gradient and exhibited a significant positive association with species richness. After statistical re-evaluation, only linear relationships were retained (p < 0.05). The results demonstrate that flow regime is a more reliable predictor of riparian vegetation zonation and functional diversity than flooding duration alone, emphasizing the role of hydrological variability in shaping ecosystem functioning within large regulated reservoirs.