Comparative analysis of machine learning methods for prediction of chlorophyll-a in a river with different hydrology characteristics: A case study in Fuchun River, China.

Eutrophication is a serious threat to water quality and human health, and chlorophyll-a (Chla) is a key indicator to represent eutrophication in rivers or lakes. Understanding the spatial-temporal distribution of Chla and its accurate prediction are significant for water system management. In this study, spatial-temporal analysis and correlation analysis were applied to reveal Chla concentration pattern in the Fuchun River, China. Then four exogenous variables (wind speed, water temperature, dissolved oxygen and turbidity) were used for predicting Chla concentrations by six models (3 traditional machine learning models and 3 deep learning models) and compare the performance in a river with different hydrology characteristics. Statistical analysis shown that the Chla concentration in the reservoir river segment was higher than in the natural river segment during August and September, while the dominant algae gradually changed from Cyanophyta to Cryptophyta. Moreover, air temperature, water temperature and dissolved oxygen had high correlations with Chla concentrations among environment factors. The results of the prediction models demonstrate that extreme gradient boosting (XGBoost) and long short-term memory neural network (LSTM) were the best performance model in the reservoir river segment (NSE = 0.93; RMSE = 4.67) and natural river segment (NSE = 0.94; RMSE = 1.84), respectively. This study provides a reference for further understanding eutrophication and early warning of algal blooms in different type of rivers.

Influence of anthropogenic disturbances on antibiotic resistance gene distributions along the Minjiang River in Southeast China.

River-reservoir systems have become ubiquitous among modern global aquatic environments due to the widespread construction of dams. However, little is known of antibiotic resistance gene (ARG) distributions in reservoir-river systems experiencing varying degrees of anthropogenic impacts. Here, the diversity, abundance, and spatial distribution of ARGs were comprehensively characterized along the main stem of the Minjiang River, a typical subtropic reservoir-river system in Southeast China using high-throughput quantitative PCR. A total of 252 ARG subtypes were detected from twelve sampling sites that were dominated by aac(3)-Via, followed by czcA, blaTEM, and sul1. Urban river waters (sites S9-S12) harbored more diverse ARGs than did the reservoir waters (sites S1-S7), indicating more serious antibiotic resistance pollution in areas with larger population densities. Dam construction could reduce the richness and absolute abundance of ARGs from upstream (site S7) to downstream (site S8). Urban river waters also harbored a higher proportion of mobile genetic elements (MGEs), suggesting that intensive human activities may promote ARG horizontal gene transfers. The mean relative abundance of Proteobacteria that could promote antibiotic resistance within microbial communities was also highest in urban river waters. Variance partitioning analysis indicated that MGEs and bacterial communities could explain 67.33%, 44.7%, and 90.29% of variation in selected ARGs for the entire watershed, aquaculture waters, and urban river waters, respectively. These results further suggest that urban rivers are ideal media for the acquisition and spread of ARGs. These findings provide new insights into the occurrence and potential mechanisms determining the distributions of ARGs in a reservoir-river system experiencing various anthropogenic disturbances at the watershed scale.

Coupled reservoir-river systems: Lessons from an integrated aquatic ecosystem assessment.

Sustainable reservoir-river management requires balancing complex trade-offs and decision-making to support both human water demands and ecological function. Current numerical simulation and optimization algorithms can guide reservoir-river operations for optimal hydropower production, irrigation, nutrient management, and municipal consumption, yet much less is known about optimization of associated ecosystems. This ten-year study demonstrates an ecosystem assessment approach that links the environmental processes to an ecosystem response in order to evaluate the impact of climatic forcing and reservoir operations on the aquatic ecosystems of a coupled headwater reservoir-river system. The approach uses a series of numerical, statistical, and empirical models to explore reservoir operational flexibility aimed at improving the environmental processes that support aquatic ecosystem function. The results illustrate that understanding the seasonal biogeochemical changes in reservoirs is critical for determining environmental flow releases and the ecological trajectory of both the reservoir and river systems. The coupled models show that reservoir management can improve the ecological function of complex aquatic ecosystems under certain climatic conditions. During dry hydrologic years, the high post-irrigation release can increase the downstream primary and macroinvertebrate production by 99% and 45% respectively. However, this flow release would reduce total fish biomass in the reservoir by 16%, providing management tradeoffs to the different ecosystems. Additionally, low post-irrigation flows during the winter season supports water temperature that can maintain ice cover in the downstream river for improved ecosystem function. The ecosystem assessment approach provides operational flexibility for large infrastructure, supports transparent decision-making by management agencies, and facilitates framing of environmental legislation.