Species extinctions are accelerating globally, yet the mechanisms that maintain local biodiversity remain poorly understood. The extinction of species that feed on or are fed on by many others (i.e. ‘hubs’) has traditionally been thought to cause the greatest threat of further biodiversity loss. Very little attention has been paid to the strength of those feeding links (i.e. link weight) and the prevalence of indirect interactions. Here, we used a dynamical model based on empirical energy budget data to assess changes in ecosystem stability after simulating the loss of species according to various extinction scenarios. Link weight and/or indirect effects had stronger effects on food‐web stability than the simple removal of ‘hubs’, demonstrating that both quantitative fluxes and species dissipating their effects across many links should be of great concern in biodiversity conservation, and the potential for ‘hubs’ to act as keystone species may have been exaggerated to date.

The modified sigmoidal bacteria growth functions (the modified Gompertz, logistic, and Richards) were used to evaluate the methane production process kinetics of agricultural wastes. The mesophilic anaerobic co-digestion experiments were conducted with various agricultural wastes as feedstocks, including cow manure, corn straw, grape leaves, vines, wine residue, strawberry leaves, and tomato leaves. The results showed that anaerobic co-digestion of cow manure and other agricultural wastes increased the methane yields while it prolonged the lag phase time. Compared with the modified Gompertz and logistic models, the modified Richards model obtained higher correlation coefficients and was able to fit experimental data better. The results of this study were expected to determine a suitable model to simulate and study the kinetic process of anaerobic co-digestion with mixed agricultural wastes as feedstocks.

The relationship between biodiversity and ecosystem functioning is a central issue in ecology, but how this relationship is affected by nutrient stress is still unknown. In this study, we analyzed the phytoplankton diversity effects on community biomass and stability along nutrient gradients in an artificial eutrophic lake. Four nutrient gradients, varying from slightly eutrophic to highly eutrophic states, were designed by adjusting the amount of polluted water that flowed into the lake. Mean phytoplankton biomass, species richness, and Shannon diversity index all showed significant differences among the four nutrient gradients. Phytoplankton community biomass was correlated with diversity (both species richness and Shannon diversity index), varying from positive to negative along the nutrient gradients. The influence of phytoplankton species richness on resource use efficiency (RUE) also changed from positive to negative along the nutrient gradients. However, the influence of phytoplankton Shannon diversity on RUE was not significant. Both phytoplankton species richness and Shannon diversity had a negative influence on community turnover (measured as community dissimilarity), i.e., a positive diversity–stability relationship. Furthermore, phytoplankton spatial stability decreased along the nutrient gradients in the lake. With increasing nutrient concentrations, the variability (standard deviation) of phytoplankton community biomass increased more rapidly than the average total biomass. Results in this study will be helpful in understanding the phytoplankton diversity effects on ecosystem functioning and how these effects are influenced by nutrient conditions in aquatic ecosystems.

In the Duliujian River, 12 water environmental parameters corresponding to 45 sampling sites were analyzed over four seasons. With a statistics test (Spearman correlation coefficient) and multivariate statistical methods, including cluster analysis (CA) and principal components analysis (PCA), the river water quality temporal and spatial patterns were analyzed to evaluate the pollution status and identify the potential pollution sources along the river. CA and PCA results on spatial scale revealed that the upstream was slightly polluted by domestic sewage, while the upper-middle reach was highly polluted due to the sewage from feed mills, furniture and pharmaceutical factories. The middle-lower reach, moderately polluted by sewage from textile, pharmaceutical, petroleum and oil refinery factories as well as fisheries and livestock activities, demonstrated the water purification role of wetland reserves. Seawater intrusion caused serious water pollution in the estuary. Through temporal CA, the four seasons were grouped into three clusters consistent with the hydrological mean, high and low flow periods. The temporal PCA results suggested that nutrient control was the primary task in mean flow period and the monitoring of effluents from feed mills, petrochemical and pharmaceutical factories is more important in the high flow period, while the wastewater from domestic and livestock should be monitored carefully in low flow periods. The results may provide some guidance or inspiration for environmental management.